2022 Volume 56 Issue 6 Pages 240-249
To understand seawater properties, such as water mass structure and mixing, geochemical analyses are useful. However, geochemical datasets for seawater that fully cover coastal areas of Hokkaido, North Japan are lacking. Here we report comprehensive geochemical analyses of seawater (salinity, δ18O, δD, and Δ14C) collected in August–September 2021 from coastal areas of Hokkaido as well as the west coast of Tohoku (Northeast Japan). These datasets are expected to improve our understanding of seawater properties around Hokkaido, thereby contributing to oceanography, climatology, biogeochemical cycles, and fishery science.
Geochemical analyses of seawater provide important insight into water properties such as water mass structure and mixing. Around Hokkaido, the Tsushima Warm Current originating from the Kuroshio Current flows northward over the Japan Sea, and after passing through the Soya Strait, it changes into the Soya Warm Current and flows eastward along the coast (Fig. 1). On the southern coast of Hokkaido, the Oyashio Current originating from the North Pacific Ocean flows westward along the coast and collides and mixes with the Tsugaru Warm Current (the Tsushima Warm Current after passing through the Tsugaru Strait) (Fig. 1). However, this is only a rough illustration of the ocean current system, and there is substantial, highly complex variation on long and short time scales. A better understanding of seawater properties around Hokkaido is needed for applications in oceanography, climatology, biogeochemical cycles, and fishery science.
Geographic map of the study area and results of a geochemical analysis of coastal seawater of Hokkaido and the west coast of North Japan. (a) Geography of the study area with major water current systems (solid line with arrows: warm currents, dashed lines with arrows: cold currents). (b) Locations of seawater sampling points with site numbers (Site No. 1–12 were visited August 20–22, 2021 and site No. 13–66 were visited September 7–16, 2021). Analyses of (c) salinity, (d) δ18O, (e) δD, and (f) Δ14C. Black/gray circles indicate cases with salinity higher/lower than 32.5 psu.
Among measurable parameters of seawater, salinity and oxygen and hydrogen stable isotopes (δ18O and δD) can be analyzed relatively easily and thus are widely used as water mass proxies.
In tandem with temperature, salinity is a fundamental indicator of seawater properties. Specifically, T-S diagrams and salinity are used to detect the mixing of different water masses and the evaporation/precipitation balance (Kodaira et al., 2016). In coastal areas, salinity can reveal mixing between seawater and freshwater (river water, groundwater, etc.) (Kubota et al., 2018a).
Similar to salinity, δ18O and δD are important parameters to characterize seawater properties such as water mass difference (Voelker et al., 2015; LeGrande and Schmidt, 2006). Also, they are particularly powerful tracers in coastal environments (e.g., estuary, delta, watershed), because isotopic compositions are distinctly different between seawater and freshwater (Nair et al., 2015; Carreira et al., 2014). In particular, the oxygen isotope ratio of calcium carbonate (CaCO3) skeletons and shells biogenically precipitated in seawater is widely used as a thermometer, as isotopic fractionation is dependent on temperature (note that this applies only when the oxygen isotopic ratio of seawater [δ18OSW] does not vary with time or can be estimated) (Kubota et al., 2015, 2017; Suzuki et al., 2005; Schöne et al., 2004; McConnaughey, 1989; Grossman and Ku, 1986). Conversely, when temperature is reconstructed by an independent method (e.g., Sr/Ca thermometers based on coral skeletons and Mg/Ca thermometers based on foraminifera shells), past δ18OSW can be quantitatively reconstructed, enabling the reconstruction of the evaporation/precipitation balance (and hence dry/wet conditions) in the surface ocean. This has been demonstrated using long-living coral skeletons (e.g., Cahyarini et al., 2014; Felis et al., 2009) and foraminifera shells preserved in marine sediment (Mohtadi et al., 2010; de Garidel-Thoron et al., 2007).
In addition, radiocarbon (Δ14C) is a powerful tracer in seawater (Broecker et al., 1985). Δ14C is a widely used tracer in a broad range of fields, including oceanography and fishery science (Ota et al., 2019; Larsen et al., 2018; Hirabayashi et al., 2017; Mitsuguchi et al., 2016; Kumamoto et al., 2013; Tsunogai et al., 1995). In the modern ocean, anthropogenic radiocarbon invades the natural carbon cycles (Östlund and Stuiver, 1980). Anthropogenic radiocarbon, bomb-14C, is a biproduct of atmospheric thermonuclear bomb testing in the 1950s and 1960s (the high-energy neutrons produced by the bombs fused with atmospheric 14N to produce 14C). After the Partial Test Ban Treaty (PTBT) became effective in 1963, new bomb-14C input ceased, but modern surface seawater is still being contaminated with bomb-14C. Δ14C in the surface seawater decreases by natural carbon cycle dynamics (i.e., the absorption of carbon by ocean and terrestrial ecosystems) and by dilution by anthropogenic carbon emissions, such as the combustion of fossil fuels in which 14C is totally decayed (14C Suess effect; Suess, 1953). In summary, Δ14C of the surface seawater changes over time and ideally needs to be reconstructed continuously. Δ14C measurements at offshore sites using research vessels are available (e.g., GLODAP: Global Ocean Data Analysis Project, Olsen et al., 2016); however, shipboard measurements provide only a snapshot of Δ14C and do not sufficiently capture the substantial spatiotemporal variability. In addition, it is quite difficult to obtain a continuous bomb-14C record, especially in high-latitude oceans lacking long-living reef-building coral skeleton made of calcium carbonate (CaCO3), suitable for generating a detailed record. In Japan, the northern limit of the reef-building coral distribution is Iki Island, West Japan (33°47'N, 129°43'E) (Yamano et al., 2001), and this area has been the northern limit of the bomb-14C record around Japan (Mitsuguchi et al., 2016). Kubota et al. (2018b, 2021) reported the first quasi-continuous seawater bomb-14C record of a relatively high-latitude coastal region, the Otsuchi area of the Pacific side of Tohoku (39°23'N, 141°57'E), using several long-living bivalve shells made of CaCO3 (Mercenaria stimpsoni). As M. stimpsoni is widely distributed around North Japan, including all coastal areas of Hokkaido and the west coast of Tohoku (the Japan Sea side), it may be possible to establish a bomb-14C record based on shells at various sites in North Japan. To the best of our knowledge, there are no reports of Δ14C of modern seawater on the west coast of Tohoku or in coastal areas of Hokkaido, and there are only limited reports of Δ14C of mollusk shells in the pre-bomb era (before 1950) (Kuzmin et al., 2001, 2007; Yoneda et al., 2000, 2007), used for 14C marine reservoir age estimation from 14C records.
Here we report a comprehensive geochemical dataset of seawater parameters (salinity, δ18O, δD, and Δ14C) collected in August–September 2021 from shallow water (<10 m) of all coastal areas around Hokkaido as well as the west coast of Tohoku. As the first such dataset for the region, these results have important implications for our understanding of modern properties of coastal water in the area and provide a basis for various practical applications.
Seawater was collected from all coastal areas around Hokkaido as well as the west coast of Tohoku during two field surveys during August 20–22, 2021 and September 7–16, 2021 (Fig. 1). Water samples were collected in two ways (Type A and B). In Type A sampling (for salinity, δ18O, and δD analyses), water samples were collected at beaches (water depth less than 0.3 m, Table 1). In type B sampling (for salinity, δ18O, δD, and radiocarbon analyses), a stainless-steel water sampler with a long rope and a 500 mL grass bottle (Sibata, Sōka, Japan) was used from wharves (water depth 0.5–10 m, Table 1), and the 500 mL water samples were divided into storage bottles according to measurement purposes. For all samples, water temperature, conductivity, and pH were measured just after sample collection by using a portable water quality analyzer (LAQUA D-200 series; Horiba, Kyoto, Japan). The dissolved oxygen (DO) concentration was measured by using a portable oxygen meter (DO–5510 HA; Lutron, Coopersburg, PA). The weather at the time of sampling was sunny in most cases, with occasional cloudy days and one rainy day on August 22, 2021 (Table 1).
| Site No. | Date | Weather | Longitude | Latitude | Water Depth (m) | Sampling Type | pH | Conductivity (mS/cm) | DO (mg/L) | Water Temperature (℃) | Salinity (psu) | Δ14C (‰) | Δ14C_stdev | δ18O (‰) | δ18O_stdev | δD (‰) | δD_stdev | Remark |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 2021/8/20 | Sunny | 144.3878 | 42.9675 | 3.5 | B | 7.62 | 51.7 | 6.6 | 19.3 | 33.9 | –41 | 5 | –0.8 | 0.2 | –4.7 | 0.7 | |
| 2 | 2021/8/20 | Sunny | 144.3892 | 42.9689 | 0.3 | A | 7.89 | 50.9 | 10.1 | 21.6 | 32.9 | –1.1 | 0.2 | –6.0 | 0.6 | |||
| 3 | 2021/8/20 | Sunny | 143.3214 | 42.2950 | 7.5 | B | 7.80 | 49.7 | 15.9 | 17.8 | 32.2 | –37 | 5 | –1.2 | 0.1 | –7.2 | 0.3 | |
| 4 | 2021/8/21 | Sunny | 143.2183 | 41.9464 | 1.5 | B | 8.00 | 49.2 | 13.5 | 20.3 | 32.9 | –30 | 5 | –0.9 | 0.2 | –4.5 | 0.6 | |
| 5 | 2021/8/21 | Cloudy | 143.2442 | 41.9331 | 0.2 | A | 8.32 | 49.4 | 14.2 | 19.6 | 31.9 | –1.2 | 0.1 | –7.8 | 0.4 | |||
| 6 | 2021/8/21 | Sunny | 142.4692 | 42.2808 | 0.3 | A | 8.40 | 49.8 | 9.2 | 29.1 | 34.1 | –0.5 | 0.1 | –2.0 | 0.6 | |||
| 7 | 2021/8/21 | Sunny | 142.3678 | 42.3233 | 3.5 | B | 8.11 | 50.9 | 10.7 | 22.3 | 34.6 | 7 | 5 | –0.3 | 0.2 | –0.7 | 0.5 | |
| 8 | 2021/8/21 | Sunny | 141.7272 | 42.6253 | 0.2 | A | 8.10 | 45.8 | 10.7 | 25.4 | 29.8 | –1.3 | 0.1 | –9.0 | 0.6 | |||
| 9 | 2021/8/21 | Sunny | 141.3111 | 42.5153 | 10.0 | B | 7.88 | 51.3 | 14.5 | 21.3 | 33.9 | 5 | 5 | –0.4 | 0.3 | –0.9 | 0.5 | |
| 10 | 2021/8/22 | Rainy | 140.9425 | 42.3417 | 1.6 | B | 7.84 | 47.2 | 8.8 | 21.1 | 32.9 | –18 | 5 | –1.1 | 0.0 | –5.1 | 0.3 | |
| 11 | 2021/8/22 | Rainy/Cloudy | 141.0228 | 42.3314 | 0.1 | A | 7.89 | 48.1 | 4.4 | 21.2 | 33.4 | –0.8 | 0.1 | –3.5 | 0.4 | |||
| 12 | 2021/8/22 | Rainy | 140.6203 | 42.5794 | 0.0 | A | 7.97 | 46.8 | 8.6 | 22.5 | 31.0 | –1.2 | 0.1 | –7.7 | 0.7 | |||
| 13 | 2021/9/7 | Sunny | 139.8589 | 39.8856 | 1.3 | B | 7.95 | 46.0 | 5.9 | 25.8 | 30.2 | 0 | 5 | –1.1 | 0.2 | –7.4 | 0.9 | |
| 14 | 2021/9/7 | Sunny | 139.8158 | 39.8572 | 0.3 | A | 8.31 | 46.0 | 7.1 | 28 | 30.9 | –1.0 | 0.1 | –6.7 | 0.6 | |||
| 15 | 2021/9/7 | Sunny | 139.9592 | 40.0981 | 0.1 | A | 8.17 | 50.7 | 8.8 | 26.9 | 33.2 | –0.7 | 0.1 | –2.2 | 0.5 | |||
| 16 | 2021/9/7 | Sunny | 139.9481 | 40.4133 | 0.2 | A | 8.26 | 51.0 | 8.8 | 25.5 | 33.4 | –0.5 | 0.1 | –1.4 | 0.5 | |||
| 17 | 2021/9/7 | Sunny | 139.8611 | 40.6081 | 2.0 | B | 8.10 | 50.5 | 5.5 | 24.9 | 34.1 | 9 | 5 | –0.4 | 0.0 | –1.0 | 0.4 | |
| 18 | 2021/9/8 | Cloudy | 139.9283 | 40.6536 | 0.2 | A | 7.78 | 52.0 | 9.1 | 21.4 | 34.1 | –0.2 | 0.1 | –0.8 | 0.3 | |||
| 19 | 2021/9/8 | Cloudy | 140.2233 | 40.7811 | 0.5 | B | 8.11 | 50.9 | 9 | 23.1 | 34.0 | 18 | 5 | –0.4 | 0.0 | –0.9 | 0.3 | |
| 20 | 2021/9/8 | Cloudy | 140.2236 | 40.7811 | 0.2 | A | 7.93 | 50.3 | 8.7 | 22.9 | Sample lost | |||||||
| 21 | 2021/9/8 | Cloudy | 140.2797 | 41.1203 | 3.6 | B | 7.99 | 51.0 | 4.8 | 22.8 | 35.0 | 22 | 5 | –0.4 | 0.1 | –0.9 | 0.1 | |
| 22 | 2021/9/8 | Cloudy | 140.3081 | 41.1025 | 0.3 | A | 8.13 | 50.9 | 8.9 | 22.8 | 34.7 | –0.5 | 0.1 | –1.3 | 0.5 | |||
| 23 | 2021/9/9 | Cloudy | 141.1833 | 41.8131 | 0.3 | A | 7.78 | 52.1 | 6.2 | 22.7 | 33.9 | –0.5 | 0.1 | –1.4 | 0.2 | |||
| 24 | 2021/9/9 | Sunny | 141.1689 | 41.8225 | 4.5 | B | 7.96 | 51.4 | 9.5 | 21.8 | 34.3 | 15 | 5 | –0.5 | 0.1 | –2.1 | 0.3 | |
| 25 | 2021/9/9 | Sunny | 140.8439 | 42.0214 | 0.0 | A | 7.98 | 50.7 | 9.7 | 23 | 33.7 | –0.7 | 0.1 | –3.2 | 0.5 | |||
| 26 | 2021/9/9 | Cloudy | 140.8272 | 42.0322 | 2.1 | B | 7.78 | 51.3 | 9.4 | 23.1 | 33.9 | –15 | 7 | –0.7 | 0.1 | –3.5 | 0.3 | |
| 27 | 2021/9/9 | Sunny | 140.5358 | 41.7083 | 2.9 | B | 8.06 | 51.8 | 8.1 | 24.2 | 34.2 | 9 | 7 | –0.4 | 0.1 | –0.8 | 0.2 | |
| 28 | 2021/9/9 | Sunny | 140.5339 | 41.7081 | 0.0 | A | 8.04 | 52.2 | 9.2 | 25.9 | 34.7 | –0.4 | 0.2 | –0.9 | 0.3 | |||
| 29 | 2021/9/10 | Sunny | 140.1986 | 41.3978 | 0.2 | A | 7.86 | 53.5 | 9.6 | 25.3 | 35.9 | –0.2 | 0.1 | 0.5 | 0.6 | |||
| 30 | 2021/9/10 | Sunny | 140.0906 | 41.4211 | 4.0 | B | 7.89 | 52.8 | 10.2 | 24.7 | 34.6 | 19 | 7 | –0.3 | 0.1 | –0.2 | 0.3 | |
| 31 | 2021/9/10 | Sunny | 140.1156 | 41.8667 | 0.2 | A | 7.98 | 52.6 | 5.8 | 28.1 | 34.8 | –0.3 | 0.0 | –0.2 | 0.4 | |||
| 32 | 2021/9/10 | Sunny | 140.1031 | 42.0100 | 1.6 | B | 7.97 | 52.1 | 10.1 | 25.5 | 35.1 | 22 | 7 | –0.3 | 0.1 | 0.2 | 0.3 | |
| 33 | 2021/9/10 | Sunny | 139.8756 | 42.2117 | 0.0 | A | 7.99 | 52.0 | 10.4 | 27 | 34.0 | –0.2 | 0.0 | 0.7 | 0.3 | |||
| 34 | 2021/9/10 | Sunny | 139.8492 | 42.4583 | 0.2 | A | 7.99 | 50.3 | 9.6 | 26.1 | 32.3 | –0.4 | 0.1 | –2.1 | 0.3 | |||
| 35 | 2021/9/10 | Sunny | 139.8472 | 42.4522 | 3.2 | B | 7.93 | 52.4 | 9.9 | 24.5 | 33.0 | 24 | 7 | –0.3 | 0.1 | 0.1 | 0.4 | |
| 36 | 2021/9/11 | Sunny | 140.2400 | 42.7756 | 0.2 | A | 7.99 | 47.4 | 8.2 | 25.2 | 30.6 | –0.6 | 0.0 | –3.3 | 0.5 | |||
| 37 | 2021/9/11 | Sunny | 140.4533 | 43.3422 | 1.6 | B | 8.10 | 51.5 | 9.5 | 24.9 | 32.5 | 20 | 7 | –0.2 | 0.0 | 0.6 | 0.3 | |
| 38 | 2021/9/11 | Sunny | 140.4608 | 43.3461 | 0.2 | A | 8.13 | 49.9 | 8.9 | 25.8 | 31.9 | –0.3 | 0.1 | –1.0 | 0.4 | |||
| 39 | 2021/9/11 | Sunny | 140.7875 | 43.1992 | 0.1 | A | 8.18 | 48.5 | 8.1 | 27.9 | 31.0 | –0.7 | 0.2 | –4.1 | 0.4 | |||
| 40 | 2021/9/13 | Sunny | 141.3853 | 43.5878 | 0.0 | A | 8.19 | 49.7 | 8.7 | 23 | 31.9 | –0.5 | 0.0 | –2.4 | 0.7 | |||
| 41 | 2021/9/13 | Cloudy | 141.5331 | 43.8525 | 3.2 | B | 8.03 | 50.4 | 9.5 | 21.5 | 32.3 | 18 | 7 | –0.1 | 0.1 | –0.3 | 0.1 | |
| 42 | 2021/9/13 | Sunny | 141.6539 | 44.1372 | 0.0 | A | 8.00 | 50.2 | 8.9 | 22.2 | 34.5 | –0.3 | 0.1 | –0.5 | 0.5 | |||
| 43 | 2021/9/13 | Sunny | 141.7744 | 44.5653 | 1.0 | B | 7.60 | 47.6 | 9.9 | 22.4 | 33.2 | 13 | 7 | –0.3 | 0.1 | –0.3 | 0.7 | |
| 44 | 2021/9/13 | Sunny | 141.7822 | 44.7156 | 0.1 | A | 7.69 | 36.5 | 9.7 | 20.4 | 25.1 | –2.8 | 0.2 | –17.8 | 0.6 | |||
| 45 | 2021/9/13 | Sunny | 141.6553 | 45.3717 | 0.0 | A | 7.82 | 50.1 | 9.8 | 19.6 | 34.7 | –0.3 | 0.1 | –0.1 | 0.3 | |||
| 46 | 2021/9/13 | Sunny | 141.6136 | 45.3097 | 3.4 | B | 7.80 | 49.8 | 9.3 | 19.5 | 34.4 | 2 | 7 | –0.2 | 0.1 | 0.0 | 0.2 | |
| 47 | 2021/9/14 | Sunny | 141.9375 | 45.5225 | 0.2 | A | 7.92 | 50.5 | 10 | 21.1 | 33.8 | –0.3 | 0.0 | –0.5 | 0.1 | |||
| 48 | 2021/9/14 | Sunny | 141.9461 | 45.5208 | 4.4 | B | 7.85 | 51.6 | 10.4 | 24.1 | 35.0 | 18 | 7 | –0.2 | 0.1 | –0.1 | 0.5 | |
| 49 | 2021/9/14 | Sunny | 142.1686 | 45.3411 | 2.5 | B | 7.46 | 49.6 | 5 | 21.4 | 34.1 | 19 | 7 | –0.4 | 0.1 | –1.6 | 0.2 | |
| 50 | 2021/9/14 | Sunny | 142.4017 | 45.1250 | 0.0 | A | 7.95 | 51.0 | 6.9 | 20.6 | 33.8 | –0.4 | 0.0 | –1.2 | 0.1 | |||
| 51 | 2021/9/14 | Sunny | 142.3928 | 45.1336 | 2.0 | B | 7.94 | 48.6 | 8.9 | 21.2 | 33.4 | 16 | 7 | –0.6 | 0.1 | –3.1 | 0.5 | |
| 52 | 2021/9/15 | Sunny | 142.8161 | 44.7097 | 2.1 | B | 8.04 | 49.6 | 9.2 | 20.7 | 33.4 | –23 | 7 | –0.5 | 0.2 | –2.7 | 0.4 | |
| 53 | 2021/9/15 | Sunny | 142.8183 | 44.7078 | 0.2 | A | 8.04 | 50.5 | 8.5 | 22.1 | 33.4 | –0.4 | 0.1 | –1.3 | 0.2 | |||
| 54 | 2021/9/15 | Sunny | 143.3742 | 44.3358 | 6.0 | B | 8.07 | 49.6 | 8.9 | 21.3 | 33.8 | 16 | 7 | –0.5 | 0.0 | –2.0 | 0.7 | |
| 55 | 2021/9/15 | Sunny | 143.3783 | 44.3278 | 0.0 | A | 7.91 | 33.4 | 9.1 | 19 | 19.8 | –4.8 | 0.0 | –33.2 | 0.4 | |||
| 56 | 2021/9/15 | Sunny | 144.0742 | 44.1236 | 0.0 | A | 8.08 | 47.8 | 9.7 | 21 | 31.4 | –0.9 | 0.1 | –5.2 | 0.3 | |||
| 57 | 2021/9/15 | Sunny | 144.1011 | 44.1278 | 3.4 | B | 7.87 | 51.0 | 9.5 | 21 | 33.3 | 24 | 7 | –0.4 | 0.1 | –1.7 | 1.1 | |
| 58 | 2021/9/16 | Sunny | 144.4525 | 43.9350 | 0.0 | A | 7.77 | 49.0 | 9.2 | 18.9 | 33.2 | –0.6 | 0.0 | –2.1 | 0.1 | |||
| 59 | 2021/9/16 | Sunny | 144.8347 | 43.9581 | 0.6 | B | 7.88 | 50.0 | 10.1 | 19.3 | 33.7 | 19 | 7 | –0.4 | 0.1 | –1.8 | 0.6 | |
| 60 | 2021/9/16 | Sunny | 144.9186 | 44.0133 | 0.0 | A | 7.97 | 48.9 | 6.7 | 20.9 | 31.3 | –0.8 | 0.1 | –4.8 | 0.5 | |||
| 61 | 2021/9/16 | Sunny | 145.2006 | 44.0200 | 5.7 | B | 8.13 | 52.0 | 10.8 | 21.3 | 33.2 | 14 | 7 | –0.4 | 0.1 | –0.5 | 0.4 | |
| 62 | 2021/9/16 | Sunny | 145.1342 | 43.9550 | 0.0 | A | 8.05 | 48.0 | 10 | 21 | Sample lost | |||||||
| 63 | 2021/9/16 | Sunny | 145.1350 | 43.6653 | 4.4 | B | 7.75 | 50.6 | 10.2 | 20.4 | 33.1 | –0.7 | 0.2 | –2.7 | 0.6 | |||
| 64 | 2021/9/16 | Sunny | 145.1367 | 43.6589 | 0.1 | A | 8.01 | 50.0 | 9.9 | 20.9 | 32.7 | –0.5 | 0.1 | –3.0 | 0.8 | |||
| 65 | 2021/9/16 | Sunny | 145.2883 | 43.3886 | 3.0 | B | 7.83 | 49.9 | 10.9 | 18.3 | 32.4 | 19 | 7 | –1.7 | 1.6 | –6.4 | 3.2 | |
| 66 | 2021/9/16 | Sunny | 145.2711 | 43.4147 | 0.0 | A | 7.97 | 47.3 | 10.3 | 18.4 | 33.5 | –0.6 | 0.0 | –3.8 | 0.0 |
All water samples were filtered at the water collection sites through a 0.45-μm Millipore PTFE membrane (Merck Millipore, Billerica, MA) using a filtration device equipped with a hand vacuum pump. For salinity, δ18O, and δD analyses, the samples were stored in 250 mL PP bottles (AsOne, Osaka, Japan) without headspace. These samples were refrigerated until analysis in a laboratory. For the radiocarbon analysis, samples were stored in 250 mL PAN (acrylonitrile butadiene methyl acrylate) bottles with a high-performance gas barrier (Nikko Hansen, Osaka, Japan), which enables long-term storage of water samples without contaminating modern carbon (Takahashi et al., 2019). The samples were shipped to the laboratory at Nagoya University and 14C in seawater dissolved inorganic carbon (DIC) was quickly analyzed using a headspace method (described later).
In a laboratory of Kobe University, salinity of the samples was measured using a portable salinity meter (LAQUA act; Horiba) calibrated with standard solutions. The measurements were conducted over two consecutive days (October 21–22, 2021). The salinity estimates of IAPSO standard seawater (Batch: P153; OSIL, Havant, United Kingdom; certified salinity of 34.994 psu; Bacon et al., 2007) were 33.1 and 34.3 psu on October 21, 2021 and October 22, 2021, respectively. These values were used as a calibration coefficient for each day.
After measuring salinity, the remaining water samples stored in PP bottles were shipped to the Atmosphere and Ocean Research Institute, The University of Tokyo, for δ18O and δD measurements using a wavelength-scanned cavity ring-down spectroscopy isotopic water analyzer (L2120-i; Picarro, Santa Clara, CA). Repeated analyses of Milli-Q water yielded values of –9.71 ± 0.06‰ (1 standard deviation [SD]) for δ18O and –65.44 ± 0.51‰ for δD. These standard deviations were used as the analytical uncertainty for isotopic measurements. Isotopic values of water samples are reported relative to Vienna Standard Mean Ocean Water (VSMOW).
The 14C in DIC was determined by a headspace method (Takahashi et al., 2021). Briefly, CO2 was extracted from seawater into the headspace of the reaction container. Then, CO2 was introduced into the vacuum line by gas expansion and cryogenically purified using an ethanol-slush trap (ca. –100°C). The purified CO2 was converted to graphite by a Fe catalyst at 620°C for 6 h. The target graphite samples were measured with a Tandetron accelerator mass spectrometer (AMS; High Voltage Engineering Europa, Amersfoort, the Netherlands) installed at the Institute for Space–Earth Environmental Research, Nagoya University. Results are reported according to the standardized system proposed by Stuiver and Polach (1977). Data were corrected for mass fractionation using the δ13C value and are presented as Δ14C in which years of formation were set to the year of collection, 2021. The error of the Δ14C analysis at Nagoya University was smaller than 8‰ (Table 1).
We report geochemical datasets for 66 sites. Note that Type A sampling and Type B sampling were often conducted at the same locations (Fig. 1, Table 1). The sampling distribution depended on whether each method was possible at the site. At two sites (Site Nos. 20 & 62), collected water samples were measured in situ and were later lost during transport. Water temperature, salinity, pH, conductivity, DO, δ18O, δD, and Δ14C values for the samples ranged from 17.8°C to 29.1°C (samples may have warmed after collection), 19.8 to 35.9 psu, 7.46 to 8.40, 33.4 to 53.5 mS/cm, 4.4 to 15.9 mg/L, –4.8‰ to –0.2‰, –33.2‰ to 0.7‰, and –41‰ to 24‰, respectively (Table 1).
We found that the Type A sampling method (shallow water sampling at beaches) tended to yield lower salinity, δ18O, and δD values with higher variability. For example, 13 of 34 water samples collected by the Type A method had salinity values of <32.5 psu (Table 2), which can likely be explained by terrestrial water due to the shallow sampling depth. Only 4 of 30 water samples collected by the Type B method (relatively deeper water sampling at wharves) resulted in salinity values of <32.5 psu. After the removal of samples with low salinity, there were no statistically significant differences in salinity, δ18O, and δD values between sampling methods (p = 0.63, 0.70, and 0.90, respectively) (Table 2).
| All data | Sampling Type A | Sampling Type B | ||||||
|---|---|---|---|---|---|---|---|---|
| Salinity | δ18O | δD | Δ14C | Salinity | δ18O | δD | Δ14C | |
| Average | 32.4 | –0.78 | –4.10 | 33.5 | –0.53 | –2.07 | 6.3 | |
| Standard deviation | 3.0 | 0.87 | 6.25 | 1.0 | 0.35 | 2.24 | 19.0 | |
| Count | 34 | 34 | 34 | 30 | 30 | 30 | 29 | |
| Data with high salinity (>32.5) | Sampling Type A | Sampling Type B | ||||||
| Salinity | δ18O | δD | Δ14C | Salinity | δ18O | δD | Δ14C | |
| Average | 33.9 | –0.48 | –1.63 | 33.8 | –0.45 | –1.57 | 7.1 | |
| Standard deviation | 0.8 | 0.23 | 1.60 | 0.7 | 0.21 | 1.59 | 17.8 | |
| Count | 21 | 21 | 21 | 26 | 26 | 26 | 26 | |
Units are psu for salinity and permil for δ18O, δD, and Δ14C.
We observed a statistically significant positive correlation between salinity and δ18O (r = 0.89, p < 0.05) as well as between salinity and δD (r = 0.92, p < 0.05) for all datapoints (Fig. 2, Table 3). When we removed water samples with low salinity (<32.5 psu), a weak correlation was still detected (r = 0.51, p < 0.05 for both combinations). There was no statistically significant correlation between salinity and Δ14C (r = 0.24, p = 0.21, Fig. 2), even when low salinity data (S < 32.5 psu) were excluded from the calculation (r = 0.24, p = 0.24, Table 3). Similarly, there were statistically significant positive correlations between δ18O and Δ14C (r = 0.48, p < 0.05) as well as δD and Δ14C (r = 0.64, p < 0.05) (Fig. 2, Table 3). The correlations became stronger when low salinity data were removed (r = 0.74 for the correlation between δ18O and Δ14C, r = 0.77 for the correlation between δD and Δ14C) (Table 3). There was no consistent pattern at sites where low salinity (S < 32.5 psu) was observed (Fig. 1). It is unlikely that this low salinity was due to rain because the weather was sunny nearly every day during the sampling campaigns. Typically, water from the west coast of North Japan had higher δ18O, δD, and Δ14C values than those of water from the Pacific side of Hokkaido (Fig. 1).
Cross-plots of salinity, δ18O, δD, and Δ14C. Cross-plots of salinity versus (a) δ18O, (b) δD, and (c) Δ14C. Cross-plots of Δ14C versus (d) δ18O and (e) δD. Black squares/gray dots are indicate cases with salinity higher/lower than 32.5 psu. The dashed line in each figure shows the fitted linear regression line for all data points.
| All data | δ18O-S | δD-S | Δ14C-S | δ18O-Δ14C | δD-Δ14C |
|---|---|---|---|---|---|
| r | 0.89 | 0.92 | 0.24 | 0.48 | 0.64 |
| R2 | 0.80 | 0.85 | 0.06 | 0.23 | 0.41 |
| N | 64 | 64 | 29 | 29 | 29 |
| t | 15.8 | 18.9 | 1.3 | 2.9 | 4.3 |
| p | 0.00 | 0.00 | 0.21 | 0.01 | 0.00 |
| Statistical significance | * | * | * | * | |
| Data with high salinity (>32.5 psu) | δ18O-S | δD-S | Δ14C-S | δ18O-Δ14C | δD-Δ14C |
| r | 0.51 | 0.51 | 0.24 | 0.74 | 0.77 |
| R2 | 0.26 | 0.26 | 0.06 | 0.54 | 0.60 |
| N | 47 | 47 | 25 | 25 | 25 |
| t | 4.0 | 3.9 | 1.2 | 5.2 | 5.9 |
| p | 0.00 | 0.00 | 0.24 | 0.00 | 0.00 |
| * | * | * | * | ||
| Data with low salinity (<32.5 psu) | δ18O-S | δD-S | Δ14C-S | δ18O-Δ14C | δD-Δ14C |
| r | 0.92 | 0.95 | 0.07 | 0.18 | 0.52 |
| R2 | 0.85 | 0.91 | 0.00 | 0.03 | 0.27 |
| N | 17 | 17 | 4 | 4 | 4 |
| t | 9.3 | 12.1 | 0.1 | 0.3 | 0.9 |
| p | 0.00 | 0.00 | 0.93 | 0.82 | 0.48 |
| * | * |
Asterisks indicate significant correlations (p < 0.05).
We reported the salinity, δ18O, δD, and Δ14C of seawater collected from shallow water of the west coast of North Japan and all coastal areas of Hokkaido. Below, we provide rough interpretations of these new data. It is important to note that deeper analyses of the observed correlations in these geochemical data are needed.
After removing water samples with low salinity, we still detected significant correlations between salinity and both δ18O and δD, suggesting that these parameters are useful for distinguishing between the Kuroshio-sourced water mass (that is, the Tsushima Warm Current and Souya Warm Current) and the Oyashio-sourced water mass. These results are consistent with the recent observation that Kuroshio has high salinity and isotopically higher δ18O and δD values due to high levels of evaporation against precipitation (salinity increases and water with heavier isotopes tends to remain during evaporation at the sea surface). These findings are also consistent with the observation that the Oyashio has lower salinity and isotopically lighter δ18O and δD values due to an inflow of terrestrial water from the surrounding continents into the North Pacific (terrestrial water has zero salinity and originates from rain and/or snow, which has a lighter isotopic composition than that of seawater).
Different Δ14C values between the Sea of Japan coast and Pacific coast likely reflect water mass differences (i.e., Kuroshio/Oyashio waters show higher/lower values). Strong correlations between δ18O and Δ14C as well as δD and Δ14C still exist if low salinity data are removed. It is unlikely that this relationship can be explained by terrestrial water (the DIC concentration of seawater is much higher than that of terrestrial water). The correlations are more consistent with the fact that Kuroshio, which is a part of the North Pacific subtropical gyre, has higher Δ14C values because emitted bomb-14C accumulated in the mixed layer of the subtropical gyre. This is also consistent with the fact that in the North Pacific, where the Oyashio Current originated, surface seawater has lower Δ14C values due to the upwelling of deep water, which is less affected by bomb-14C and shows greater 14C decay due to long-term isolation from the atmosphere (deep-sea water circulates around the globe in ~2000 years).
We thank T. Fujiki and R. Hayashi at the Okayama University of Science for their support with in situ seawater measurement and water sampling during the field survey. We also thank K. Seike at National Institute of Advanced Industrial Science and Technology for his advice on the field survey. We also thank M. Yoshioka at Kobe University for her support with salinity measurement of seawater in the laboratory. We also thank T. Miyajima at Atmosphere and Ocean Research Institute for help with analyzing hydrogen and oxygen isotopes of water samples. This study was financially supported by a Grants-in-Aid for Young Scientists from Foundation of Kinoshita Memorial Enterprise to K. Kubota, the Joint Research Program of the Institute for Space–Earth Environmental Research to K. Kubota, a Grants-in-Aid for field surveys from the Fukada Geological Institute to K. Sakai.
