Portal de Periódicos da CAPES

Oceanic iron supply mechanisms which support the spring diatom bloom in the Oyashio region, western subarctic Pacific

2011; American Geophysical Union; Volume: 116; Issue: C2 Linguagem: Inglês

10.1029/2010jc006321

2156-2202

Jun Nishioka, Tsuneo Ono, Hiroaki Saito, Keiichiro Sakaoka, Takeshi Yoshimura,

Oceanographic and Atmospheric Processes

Journal of Geophysical Research: OceansVolume 116, Issue C2 Free Access Oceanic iron supply mechanisms which support the spring diatom bloom in the Oyashio region, western subarctic Pacific Correction(s) for this article Correction to "Oceanic iron supply mechanisms which support the spring diatom bloom in the Oyashio region, western subarctic Pacific" Jun Nishioka, Tsuneo Ono, Hiroaki Saito, Keiichiro Sakaoka, Takeshi Yoshimura, Volume 116Issue C4Journal of Geophysical Research: Oceans First Published online: April 12, 2011 Jun Nishioka, Jun Nishioka [email protected] Pan-Okhotsk Research Center, Institute of Low Temperature Science, Hokkaido University, Sapporo, JapanSearch for more papers by this authorTsuneo Ono, Tsuneo Ono Hokkaido National Fisheries Research Institute, Kushiro, JapanSearch for more papers by this authorHiroaki Saito, Hiroaki Saito Touhoku National Fisheries Research Institute, Shiogama, JapanSearch for more papers by this authorKeiichiro Sakaoka, Keiichiro Sakaoka Faculty of Fisheries Science, Hokkaido University, Hakodate, JapanSearch for more papers by this authorTakeshi Yoshimura, Takeshi Yoshimura Central Research Institute of Electric Power Industry, Abiko, JapanSearch for more papers by this author Jun Nishioka, Jun Nishioka [email protected] Pan-Okhotsk Research Center, Institute of Low Temperature Science, Hokkaido University, Sapporo, JapanSearch for more papers by this authorTsuneo Ono, Tsuneo Ono Hokkaido National Fisheries Research Institute, Kushiro, JapanSearch for more papers by this authorHiroaki Saito, Hiroaki Saito Touhoku National Fisheries Research Institute, Shiogama, JapanSearch for more papers by this authorKeiichiro Sakaoka, Keiichiro Sakaoka Faculty of Fisheries Science, Hokkaido University, Hakodate, JapanSearch for more papers by this authorTakeshi Yoshimura, Takeshi Yoshimura Central Research Institute of Electric Power Industry, Abiko, JapanSearch for more papers by this author First published: 16 February 2011 https://doi.org/10.1029/2010JC006321Citations: 44AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Abstract [1] Multiyear (2003–2008) time series observations along the A line provided information on the temporal variability of the dissolved iron (diss-Fe) concentration in the Oyashio region of the western subarctic Pacific, and the data indicated that there was an annual cycle in the concentration of surface diss-Fe occurring every year. Diss-Fe was supplied into the surface water in this region every winter and supports the spring phytoplankton bloom after development of the thermocline. The diss-Fe concentration was drawn down during the phytoplankton bloom period and was depleted in summer in some water masses. Then diss-Fe increased from autumn to winter with the increasing depth of the surface mixed layer. The high diss-Fe concentrations in the surface layer in winter were controlled by mesoscale oceanic intrinsic processes, such as vertical winter mixing and horizontal Fe-rich intermediate water transport. Difference in magnitude of the winter mixing processes among different water masses caused the heterogeneous distribution of diss-Fe concentration in the surface layer. Moreover, the vertical section profiles along a cross-Oyashio transect showed the occurrence of Fe-rich intermediate water, and upward transport of materials from the intermediate water to the surface layer via tidal and winter mixing processes are important mechanisms to explain the high winter surface diss-Fe concentrations. Additionally, the substantially higher diss-Fe/NO3 ratio in the winter surface layer in this studied area other than the high-nutrient low-chlorophyll region indicates that the winter surface water in the Oyashio and the Oyashio-Kuroshio transition zone has a high potential to stimulate phytoplankton growth. 1. Introduction [2] The western subarctic Pacific (WSP) is an area of confluence of water masses that are carried by the Oyashio, Kuroshio, and mesoscale eddies. The Oyashio is the western boundary current of the northern North Pacific, which is formed by cold, fresh, nutrient-rich upwelled waters from the western subarctic gyre flowing along the eastern side of the southern Kuril islands (Figure 1a). The Kuroshio, western boundary current of the subtropical North Pacific, transports warm, saline, nutrient-poor water into the midlatitude of the North Pacific (Figure 1a). Previous studies have shown that the WSP has a larger seasonal amplitude in chlorophyll a and primary productivity than in the eastern subarctic Pacific (ESP) [Harrison et al., 2004; Chierici et al., 2006]. Phytoplankton blooms regularly occur in spring in the Oyashio and Oyashio-Kuroshio transition zone [Saito et al., 2002; Okamoto et al., 2010], which corresponds with the area of the highest biological drawdown of pCO2 and nutrient in the world ocean [Takahashi et al., 2002; Chierici et al., 2006]. Other previous studies showed that the WSP works as an effective biological pump and is an important region in the global carbon cycle [Longhurst et al., 1995; Honda, 2003; Schlitzer, 2004; Buesseler et al., 2007; Boyd et al., 2008]. Figure 1Open in figure viewerPowerPoint Schematic drawing of (a) the water structure in the western subarctic Pacific, (b) sampling stations along the observed A line and (c) ship'strack of the underway survey in January 2008 of diss-Fe measurements in the Oyashio region. OY, Oyashio; KR, Kuroshio; OSIW, Okhotsk Sea Intermediate Water; WSP, western subarctic Pacific. [3] On the other hand, despite the high biological productivity in the Oyashio and the Oyashio-Kuroshio transition zone, the whole subarctic Pacific is well known as one of the high-nutrient low-chlorophyll (HNLC) regions of the world's oceans and mesoscale iron (Fe) enrichment experiments conducted in subarctic Pacific clearly reveal that Fe limits phytoplankton growth [Tsuda et al., 2003; Boyd et al., 2004; Harrison et al., 2004]. Since Fe is an essential micronutrient for the control of phytoplankton growth in the subarctic Pacific [Martin et al., 1989], there is considerable interest in the seasonal cycle of the dissolved Fe (diss-Fe) concentration to determine the source and seasonal timing of input, which leads to the occurrence of the spring phytoplankton bloom in the Oyashio and the Oyashio-Kuroshio transition zone. [4] To date, atmospheric dust has been considered to be the most important source of Fe in this region. Previous studies indicate that there is a longitudinal dust gradient across the North Pacific, that is, the flux of dust containing Fe over the WSP is an order of magnitude higher than that in the ESP [e.g., Duce and Tindale, 1991; Mahowald et al., 2005]. This is due to the close proximity to the second largest dust source in the world, the Gobi Desert, and this has been believed to be the leading cause for the longitudinal differences in biological production between the WSP and the ESP [Harrison et al., 2004]. [5] On the other hand, recent studies have indicated that the source and the oceanic transport processes of Fe from the continental shelf are increasingly recognized as important for Fe supply to the open ocean [e.g., Croot and Hunter, 1998; Elrod et al., 2004; Johnson et al., 2005; Moore and Braucher, 2008]. In the Southern Ocean, another HNLC region, the potential contribution from multiple Fe sources including dust and oceanic Fe supply processes was evaluated quantitatively for regions adjacent to island or the Antarctic Peninsula [Planquette et al., 2007; Blain et al., 2008; Dulaiova et al., 2009] and oceanic region [Croot et al., 2004; Ellwood et al., 2008; Bowie et al., 2009]. As the Southern Ocean, oceanic Fe supply processes have reported in the subarctic Pacific [e.g., Lam et al., 2006; Nishioka et al., 2007; Lam and Bishop, 2008; Cullen et al., 2009] in addition to atmospheric dust Fe supply. Therefore, we need more detailed investigations in order to clarify the dominant source of Fe which supports biological production in the subarctic Pacific. [6] The temporal variability in the Fe concentration provides important information for determining Fe sources and has been studied in other regions of the coastal and open ocean surface of the North Pacific [e.g., Johnson et al., 1999, 2001; Elrod et al., 2008; Boyle et al., 2005; Chase et al., 2005]. The temporal variability of open ocean Fe concentrations in near surface waters in the central North Pacific indicated that the highest Fe concentrations were observed during periods of high Asian dust transport in spring and suggested the occurrence of significant interannual differences in the near surface Fe concentration responded to dust input [Boyle et al., 2005]. In the ESP, vertical profiles of diss-Fe from the Ocean Station Papa (OSP) in different seasons showed little seasonal variation in the surface diss-Fe concentration, ranging between 0.06 and 0.10 nM with one exception in February 1999 (0.23 nM) [Nishioka et al., 2001]. In the WSP, reported time series data, which only covered the winter to late spring season, showed a clear seasonal difference in the diss-Fe concentrations in the surface water [Nishioka et al., 2007]. They also reported possible Fe supply processes into the WSP through intermediate water originating from the Sea of Okhotsk, in addition to the traditional view of dust Fe supply to this region [e.g., Uematsu et al., 1983; Duce and Tindale, 1991; Jickells and Spokes, 2001; Measures et al., 2005; Buck et al., 2006]. Additionally, lateral Fe transport from continental margin around Kamchatka [Lam and Bishop, 2008] is a possible source to the WSP. To date, however, time series data which cover all seasons have not been reported and the annual cycle in Fe concentrations has not been fully examined in the WSP. [7] In this study, we report a long-term record of surface diss-Fe concentrations from 2003 to 2008 with a seasonal resolution along the monitoring observation line (A line) which crosses the Oyashio (Figure 1b) in order to detail the annual cycle of surface diss-Fe concentration. Additionally, underway surface Fe measurements along the ship's transect have been shown to be a good tool to collect Fe distribution data over a wide area in the surface layer [Vink and Measures, 2001; Johnson et al., 2003]. To elucidate the Fe supply process in this region, details of the spatial distribution of diss-Fe concentration in the winter surface (Figure 1c) were investigated by using underway autosequence sampling and analytical system, and also vertical sections of diss-Fe along the A line were measured. We then discuss the mechanism driving the annual cycle of surface diss-Fe concentration in this region. 2. Methods Time Series Diss-Fe Observation [8] The data presented here were obtained during one to eight cruises a year along the observation line of the Fisheries Research Agency (A line [Saito et al., 2002]) which crosses the Oyashio current (Figure 1b), from January 2003 to January 2008, using the R/V Hokko-Maru, Wakataka-Maru, Tankai-Maru and Oshoro-Maru. The data from January to May 2003 have previously been reported [Nishioka et al., 2007]. The cruise information in this study is shown in Table 1. From January 2003 until March 2004, samples were collected from the surface (10 m) to 800 m maximum at one to eight stations sampled regularly along the A line, using acid-cleaned Teflon coated 10 L X-Niskin sampling bottles suspended on a Kevlar line. From January 2005, acid-cleaned Teflon coated 10 L X-Niskin sampling bottles attached to a modified clean CTD-carousel multisampler system (CTD-CMS system: SBE-911 plus and SBE-32 water sampler Sea Bird Electronics Inc.), and samples were collected from 10 to 3000 m at six to nine stations along the A line. A comparison between sampling bottles on the Kevlar line and CTD-CMS was conducted during the January 2005 cruise, and no differences were observed between the sampling methods (Figure 2). The samples for diss-Fe measurement were filtered using 0.22 μm acid-cleaned filters (Millipac-100, Millipore) under gravity pressure [Nishioka et al., 2007]. Details of the sampling are described by Nishioka et al. [2001, 2007]. All filtrate samples were adjusted to pH 3.2 with 10 M formic acid-2.4 M ammonium formate buffer solution and determined with an automatic Fe (III) flow injection analytical (FIA) system (Kimoto Electric, Ltd.) using chelating resin preconcentration and chemiluminescence detection [Obata et al., 1993]. These samples were allowed to sit at pH 3.2 for more than 1 day at room temperature and measured onboard or within onshore laboratory approximately 2 weeks after each cruise. We should note that to standardize the analysis methods through the time series observation period, we acidified the all samples to pH 3.2 prior to analysis in this study. The determined diss-Fe concentration is in the form of chemically labile species which are leachable Fe in 0.22 μm filtrate at pH 3.2 and strongly react with 8-hydroxyquinoline resin in the FIA chemiluminescence detection system [Obata et al., 1997]. All sample treatments were performed in a laminar flow clean air hood in the onboard laboratory. Figure 2Open in figure viewerPowerPoint Results of the comparison between sampler bottles on the Kevlar line and CTD-CMS conducted during the January 2005 cruise. Error bars represent relative standard deviation within 5% for replicate measurements of a standard seawater sample during test cruise, some are smaller than symbols. Table 1. Information on Time Series Iron Observation in the Oyashio Region Along the A Linea Year Month Stations Vessel Cruise Observed Minimum and Maximum Depth (m) Sampling Method Source 2003 Jan A3, A4, A5, A7, A9, A11 A15 R/V Hokko-Maru HK0301 10–800 Kevlar wire reported by Nishioka et al. [2007] Feb A7 R/V Oshoro-Maru OS131 10–300 Kevlar wire reported by Nishioka et al. [2007] Mar A4, A7 R/V Oshoro-Maru OS133 10–300 Kevlar wire reported by Nishioka et al. [2007] Apr A4, A7, A11, A17 R/V Wakataka-Maru WK0304 10–800 Kevlar wire reported by Nishioka et al. [2007] Apr (revisit) A4, A7, A11 R/V Wakataka-Maru WK0304 10–800 Kevlar wire reported by Nishioka et al. [2007] May A4, A7, A11, A17 R/V Wakataka-Maru WK0305 10–800 Kevlar wire reported by Nishioka et al. [2007] May (revisit) A4, A7, A11 R/V Wakataka-Maru WK0305 10–800 Kevlar wire reported by Nishioka et al. [2007] Oct A3, A4, A5, A7, A9, A11, A13, A15 R/V Hokko-Maru HK0310 10 Kevlar wire 2004 Jan A3, A4, A5, A7, A9, A11 R/V Hokko-Maru HK0401 10–800 Kevlar wire Mar A3, A4, A5, A7, A9, A11 R/V Tankai-Maru TK0403 10 Kevlar wire 2005 Jan A3, A4, A5, A7, A9, A11, A13, A15 R/V Hokko-Maru HK0501 10–3000 clean CTD-CMS system vertical section observation Mar A5, A7, A9, A11, A13, A17 R/V Tankai-Maru TK0503 10 clean CTD-CMS system May A3, A4, A5, A7, A9, A11, A13, A15, A17 R/V Hokko-Maru HK0505 10–3000 clean CTD-CMS system vertical section observation July A3, A4, A5, A7, A9, A11, A13, A15 R/V Hokko-Maru HK0507 10 clean CTD-CMS system Sep A3, A4, A5, A7, A9, A11, A13, A15, A17 R/V Hokko-Maru HK0509 10 clean CTD-CMS system Dec A4, A5, A7, A9, A11, A13, A15, A17 R/V Hokko-Maru HK0512 10–3000 clean CTD-CMS system vertical section observation 2006 Jan A4, A5, A7, A9, A11, A13 R/V Hokko-Maru HK0601 10–3000 clean CTD-CMS system vertical section observation 2007 Jan A3, A4, A5, A7, A9, A11, A13, A15, A17 R/V Hokko-Maru HK0701 10 clean CTD-CMS system May A3, A4, A5, A7, A9, A11, A13 R/V Hokko-Maru HK0705 10 clean CTD-CMS system Oct A3, A4, A5, A7, A9, A11, A13, A15, A17 R/V Hokko-Maru HK0710 10 clean CTD-CMS system 2008 Jan A4, A5, A7, A9, A11, A13, A15, A17 R/V Hokko-Maru HK0801 10 clean CTD-CMS system underway observation a Year, date, stations, vessel, cruise, observed depth range, and sampling method. [9] Our Fe measurement method and reference seawater were vetted by using SAFe cruise [Johnson et al. 2007] reference standard seawater (distributed by the Moss Landing Marine Laboratory (MLML) for an intercomparison study), with our results comparing favorably for dissolved iron concentration in ∼0.1 and ∼1 nM (SAFe reference standard seawater containing 0.094 ± 0.008 nM (S) and 0.923 ± 0.029 nM (D2) iron (http://www.geotraces.org/) were measured to be 0.10 ± 0.010 nM (n = 3) and 0.99 ± 0.023 (n = 3) by our method, respectively (the reference seawater was analyzed on 26 December 2006)). [10] Nutrients and chlorophyll a concentrations were also analyzed for water samples. Nutrients concentrations were measured using a BRAN-LUEBBE autoanalyzer (TRACCS 800), and chlorophyll a concentrations were measured onboard by fluorometry (Turner Designs Model 10-AU) with nonacidified method as described by Welschmeyer [1994]. Hydrographic data was also collected at all stations using a CTD sensor. The surface mixed layer depth was calculated as the depth at which the potential density of the water column increased by 0.125 units [Levitus, 1982] relative to the sea surface. Underway Diss-Fe Observations in the Winter Surface Layer [11] In January 2008, clean surface water (1.5–4 m depth) was collected using a towed fish, metal-free sampling system [Tsumune et al., 2005]. The cruise track made a three-sided box around the A line and the ship's track during the underway observation is shown in Figure 1c. The system consists of 50 kg towed fish covered with metal-free epoxy paint and Teflon tubing (I.D. 12 mm) covered by PVC. The tubing was set through the center of the towed fish and the sample water flowed from the top of towed fish set on the side of the ship to the onboard laboratory, using an air-driven Teflon pump. The sample was filtered through a 0.22 μm cartridge type filter (OPTICAP, Millipore) at the end of the tubing, and the sample led directly into an automatic Fe (III) flow injection analytical system (Kimoto Electric, Ltd.) which has an autosequence sampling system (J. Nishioka et al., manuscript in preparation, 2011). All sample treatments were conducted using an in-line system throughout sampling to analysis. This underway autosequence sampling and analytical system enables the measurement of diss-Fe concentrations in the surface layer with a high resolution (1 sample per 15–20 min) along the ship track with a 10–15 knot ship speed (diss-Fe was measured every 2.5–4 nautical miles). The determined diss-Fe concentration is in the form of chemically labile species which are in an immediately leachable form of Fe in 0.22 μm filtrate at pH 3.2 and strongly react with 8-hydroxyquinoline resin in the FIA chemiluminescence detection system. Nitrate concentrations were also analyzed for water samples, which were collected from the end of sampling tubing every 15–30 min. 3. Results Time Series Data of Surface Diss-Fe Concentrations [12] Diss-Fe concentrations along A line were measured, and changes in surface (mean value of concentrations in the surface mixed layer or concentrations at 10 m depth) diss-Fe concentrations, surface nitrate concentrations, surface mixed layer depth (MLD), surface chlorophyll a concentrations, from January 2003 to January 2008 are shown in Figure 3 and Table S1 in the auxiliary material. Figure 3Open in figure viewerPowerPoint Changes in (a) surface diss-Fe concentrations, (b) surface nitrate concentrations, (c) surface mixed layer depth (MLD), (d) surface chlorophyll a concentrations, and (e) number of the dust events which were observed by the Japan Meteorological Agency at 69 meteorological stations throughout Japan using visually transmittance survey (http://www.data.kishou.go.jp/obs-env/kosahp/kosa_table_1.html) from January 2003 to January 2008. [13] Mineral aerosols from the Asian continent are transported to the North Pacific. High mineral dust concentrations have been frequently observed in spring and transported by the movement of a low-pressure system over Japan [Uematsu et al., 2002]. The number of dust events observed by the Japan Meteorological Agency at 69 meteorological stations throughout Japan using a visual transmittance survey (http://www.data.kishou.go.jp/obs-env/kosahp/kosa_table_1.html) is shown in Figure 3e. [14] Mean monthly compiled data (from the same data set as Figure 3 and Table S1) for the 5 year studied period of the surface diss-Fe and nitrate concentrations, and MLD along A line are shown in Figure 4. In the coastal area of this region, there is the Coastal Oyashio water which has different characteristics (identified by surface salinity <33.0 [Ohtani, 1989]; see section 3.2). The processes which control seasonality of Fe concentrations in the Coastal Oyashio water are probably different from the Oyashio and Oyashio-Kuroshio transition zone. Therefore, to avoid influence of the Coastal Oyashio water, the plotted data in the Figures 3 and 4 were selected from A4 to A15 (these stations include the Oyashio water and part of the Oyashio-Kuroshio transition zone), except A4 May 2005 which data was excluded due to clearly influence of surface flow of the coastal water. Figure 4Open in figure viewerPowerPoint Monthly compiled (a) surface diss-Fe concentration, (b) nitrate concentration, and (c) mixed layer depth along the A line in the Oyashio region. (For Figures 4a and 4b, the plots are the mean values of all concentrations in the surface mixed layer in each month which are presented in Table S1.) Error bars represent ± 1SD value. [15] This time series observations show an annual cycle of the diss-Fe concentration in the surface layer in this studied area as follows. [16] 1. In "winter," from the monthly compiled data, maximum value (0.92 ± 0.54 nM, mean ± 1SD, monthly compiled data in Figure 4a) is recorded in March when the surface mixed layer became deepest, the same as nitrate (Figure 4b). [17] 2. In "spring," a shallow thermocline was formed in the surface layer (Figure 4c) and both of the diss-Fe and nitrate concentration decreased during the spring phytoplankton bloom (Figures 3a, 3b, 3d, 4a, and 4b). It is also noteworthy that we observed some depleted values of diss-Fe concentrations at some stations along the A line in May (Figures 3a and 4a and Table S1), when high nitrate concentrations (10.25 ± 6.28 μM) still remain in the surface water (Figure 4b and Table S1). [18] 3. In "summer," diss-Fe was depleted in summer in some stations (0.10 ± 0.14 nM, mean ± 1SD, monthly compiled data in July in Figure 4a) along the observed line. [19] 4. In "autumn to winter," diss-Fe concentration increased from autumn to winter (0.61 ± 0.49 nM, mean ± 1SD, monthly compiled data in January in Figure 4a) with the increase of MLD (Figure 4c). Spatial Distribution of Diss-Fe in the Winter Surface Layer [20] The results of surface salinity, temperature, diss-Fe concentration and nitrate concentration measurements using the underway autosequence sampling analytical system conducted in winter, January 2008, are shown in Figure 5 and Table S2. The diss-Fe concentration level in the surface layer clearly changed with the changing mesoscale water hydrographic features, such as the water properties of salinity and temperature (Figures 5a, 5b, and 5c). The area we observed was occupied by Coastal Oyashio water (surface salinity 34.0 [Ono et al., 2005]), and the front water between Oyashio water and Kuroshio-oriented warm water (surface salinity 33.6–34.0). Among these, the Coastal Oyashio water has the highest diss-Fe concentration up to 2.8 nM (Figure 5c). Although the other waters had moderately high diss-Fe concentrations (mean 0.33 ± 0.18 nM; mean ± 1 SD, Figure 5c), our data showed that the center of both Oyashio water domain and Kuroshio-oriented warm water domain has relatively low diss-Fe value (∼0.2 nM). The front water generally showed relatively high diss-Fe values (0.3–0.6 nM). Figure 5Open in figure viewerPowerPoint Surface counter plot of (a) sea surface salinity, (b) temperature, (c) surface diss-Fe concentrations, and (d) nitrate concentrations during underway observations in January 2008 in the Oyashio region. COW, Coastal Oyashio water; OY, Oyashio; KR, Kuroshio. Vertical Section Fe Distribution Along A Line [21] Section profiles of temperature and salinity along the A line in January and December 2005 and January 2006 are shown in Figure 6. The section profiles of diss-Fe concentrations are shown in Figure 7. All data are shown in Table S3. The section for diss-Fe indicates that the diss-Fe concentrations in winter were substantial in the surface mixed layer (0.53 ± 0.25 nM, mean ± 1SD, mean value of all data from MLD in January 2005, December 2005, January 2006), and higher in the intermediate water (1.36 ± 0.50 nM, mean ± 1SD, mean value of all data in intermediate water (26.6–27.5 σθ) in January and December 2005 and January 2006). The core of high diss-Fe concentrations of the intermediate water is consistent with Okhotsk Sea Intermediate Water (OSIW) [Yasuda et al., 2001; Ohshima et al., 2010], which contain high concentration of Fe [Nishioka et al., 2007]. Figure 6Open in figure viewerPowerPoint Vertical section profiles of (a) temperature and (b) salinity along the A line in January 2005, December 2005, and January 2006. Figure 7Open in figure viewerPowerPoint Vertical section profiles of diss-Fe along the A line in January 2005, December 2005, and January 2006. Red dots in Figure 7 (top) indicate observed surface mixed layer depth. 4. Discussion Annual Cycle of Diss-Fe Concentration [22] Nishioka et al. [2007] reported the seasonal variation of surface diss-Fe, nitrate, MLD and chlorophyll a concentration from January to May 2003, and the pattern of seasonal change in diss-Fe concentration in the surface mixed layer was similar to that of macronutrients. These results suggested that deep winter water mixing resulted in relatively high winter concentration of Fe in the surface water of this region. The longer-term time series observations in this study clearly show an annual cycle of the diss-Fe concentration in the surface layer in this studied area, which occurred every year (Figures 3a and 4a). High diss-Fe levels in the surface mixed layer were detected every winter from 2003 to 2008, implying that Fe is supplied every winter into the surface water (Figure 3a). [23] To discuss the Fe supply processes which drive the annual cycle of the surface diss-Fe concentration, the time series diss-Fe concentration were plotted with the monthly variation in the number of dust events in Figures 3a and 3e. Dust events were rare in autumn to winter, and this is the period during which the surface diss-Fe concentration increased. Therefore, although a part of the surface diss-Fe concentrations may be affected by the spring dust Fe supply, dust events probably do not markedly contribute to the high diss-Fe concentration in the winter surface mixed layer and the annual cycle of the surface diss-Fe in the Oyashio and Oyashio-Kuroshio transition zone. [24] The diss-Fe concentrations decreased every spring with the increase in chlorophyll a concentrations, and was depleted in some stations in summer (Figures 3a and 4a and Table S1). Part of this is probably due to the biological uptake of diss-Fe during the spring phytoplankton bloom. The previously conducted mesoscale Fe enrichment experiment in the subarctic North Pacific clearly showed that artificially enriched Fe in the dissolved fraction was rapidly transformed into suspended labile particulate Fe during phytoplankton growth due to aggregate to particles and reduced Fe bioavailability [Nishioka et al., 2003]. Therefore, in addition to the biological uptake, the transformation process probably occurs also during the natural spring phytoplankton bloom. The transformation process reduces the diss-Fe concentration and its bioavailability, and may affect the termination of the bloom. Diss-Fe Supply in the Winter Surface Layer [25] We discuss here the potential sources of Fe which can explain the high diss-Fe levels in the surface mixed layer in winter. Details of the spatial distribution of diss-Fe with water mass hydrographic features in the surface are important for understanding the Fe supply processes to the ocean surface. The spatial scale of atmospheric dust events is well larger than that of oceanic intrinsic physical processes, such as eddy, jet and turbulent upwelling processes. Therefore, if atmospheric Fe input dominates the Fe supply to the surface layer, the diss-Fe concentration would not vary with water mass hydrographic features. While, if diss-Fe concentrations vary with the water mass hydrographic features, the concentrations can be considered to be controlled by oceanic processes, such as winter turbulence, vertical mixing and upwelling/downwelling of each water mass. [26] The heterogeneity of the spatial distribution of diss-Fe concentrations explains the wide range of surface water values found throughout the course of the time series study and clearly depicts an association with various water masses. The diss-Fe concentration level in the surface layer in January 2008 clearly changed with the changing mesoscale water hydrographic features, such as the thermohaline propert