The waters off southeastern Hokkaido, Japan, in the western subarctic Pacific are highly biologically productive, due to a substantial spring phytoplankton bloom. Previous studies have reported that the scale and timing (seasonality) of spring blooms are influenced by the Coastal Oyashio Water (COW), a low‑density, iron‑rich water mass. The COW is formed from relatively low-saline water of the East Sakhalin Current (ESC) and sea ice meltwater in the Okhotsk Sea; however, the detailed formation processes are still not well understood. Recent rapid warming in the waters off southeastern Hokkaido and environmental changes in the Okhotsk Sea, such as sea ice variability, may cause variations in sea surface temperature (SST) and the COW distribution, potentially influencing the spring bloom seasonality. However, the spatiotemporal relationships among SST, COW, and the spring bloom seasonality, as well as the drivers of COW variability, remain unclear. Therefore, this study aimed to clarify how SST and the COW regulate the spring bloom seasonality off southeastern Hokkaido and how environmental changes in the Okhotsk Sea influence the COW.
  We developed an algorithm to derive salinity from light absorption coefficient of chromophoric dissolved organic matter (a_CDOM) collected during the several research cruises in the northwestern subarctic Pacific from 2006 to 2023 (Fig. 1). We obtained satellite-derived SST from NASA’s Ocean Color Web (https://oceancolor.gsfc.nasa.gov/); chlorophyll-a (chl.a) concentration and a_CDOM from GlobColour (https://hermes.acri.fr/); sea ice concentration (SIC) from the University of Bremen (https://seaice.uni-bremen.de/start/); and geostrophic velocity from the Copernicus Marine Environment Monitoring Service (https://data.marine.copernicus.eu/products) around Hokkaido from 2003−2023. We estimated salinity from a_CDOM using the algorithm and then determined the water mass for each pixel using temperature-salinity diagrams (Hanawa and Mitsudera, 1987; Oguma et al., 2008). The peak chl.a concentration (chl.a_peak) and onset date of the spring bloom were calculated by fitting a Gaussian function to the chl.a concentration time series from January to August (Sasaoka et al., 2011). Four regions of interest (ROIs; Fig. 1) were defined, and mean SST, COW coverage ratio (R_COW), chl.a_peak, and onset date were calculated in each ROI. ESC strength was quantified using southward alongshore velocity near 47°N off eastern Sakhalin (red line in Fig. 1).
  In the most offshore ROI, chl.a_peak ranged from 1.52−9.07 mg m^-3 and showed a weak negative correlation (Spearman’s rank correlation) with SST (ρ = -0.44, p = 0.048), while a significant positive correlation with R_COW (ρ = 0.57, p = 0.007). These results indicate that nutrients supply from the COW evidently effects on phytoplankton activity in this ROI and subtropical warm water mass (e.g., warm core ring) with low nutrients reduce it. In the most coastal ROI, the bloom onset date varied from early March and early April and occurred climatologically when SST reached approximately 1.4 °C (median value). Therefore, we calculated the rank correlation coefficient between the bloom onset date and the date when the SST reached 1.4 ℃ and found a strong positive correlation (ρ = 0.99, p < 0.001). These results suggest that the bloom‑forming phytoplankton in the most coastal ROI have a physiological threshold for growth at around 1.4 °C below which photosynthetic activity is suppressed even when nutrients are sufficiently supplied by winter vertical mixing or the COW intrusion.
  The lag correlations between the climatological ESC strength and R_COW in the most coastal ROI indicated the strongest correlation when R_COW lagged the ESC by 8 weeks  (r = 0.82, p < 0.001). However, no clear relationship was observed between the interannual variability of ESC strength and R_COW. In contrast, the interannual relationship between R_COW and SIC showed a significant positive correlation (ρ = 0.52, p = 0.019). This result suggests that the COW formation is influenced not only by the ESC but also by sea ice and its meltwater. Sea ice meltwater in the southern Okhotsk Sea is fresh and is known to contain abundant dissolved iron and phytoplankton seeds (e.g., Kuroda et al., 2019), characteristics that are also typical of COW. Although further investigation is needed to quantify the contributions of these processes, our results suggest that warming off southeastern Hokkaido may advance the coastal spring bloom and reduce its scale in the offshore region, while interannual sea ice variability in the Okhotsk Sea may control offshore bloom scale via the extent of COW coverage.
 

Spring phytoplankton bloom,Ocean color remote sensing,Coastal Oyashio,Ocean environmental changes,Oyashio,Okhotsk Sea,Sea ice