The most problematic issue in the ocean color application is the presence of heavy clouds, especially in polar regions. For that reason, the demand for the ocean color application in polar regions is increased. As a way to overcome such issues, we conducted the reconstruction of the chlorophyll-a concentration (CHL) data using the machine learning-based models to raise the usability of CHL data. This analysis was first conducted on a regional scale and focused on the biologically-valued Cape Hallett, Ross Sea, Antarctica. Environmental factors and geographical information associated with phytoplankton dynamics were considered as predictors for the CHL reconstruction, which were obtained from cloud-free microwave and reanalysis data. As the machine learning models used in the present study, the ensemble-based models such as Random forest (RF) and Extremely randomized tree (ET) were selected with 10-fold cross-validation. As a result, both CHL reconstructions from the two models showed significant agreement with the standard satellite-derived CHL data. In addition, the reconstructed CHLs were close to the actual CHL value even where it was not observed by the satellites. However, there is a slight difference between the CHL reconstruction results from the RF and the ET, which is likely caused by the difference in the contribution of each predictor. In addition, we examined the variable importance for the CHL reconstruction quantitatively. As such, the sea surface and atmospheric temperature, and the photosynthetically available radiation have high contributions to the model developments. Mostly, geographic information appears to have a lower contribution relative to environmental predictors. Lastly, we estimated the partial dependences for the predictors for further study on the variable contribution and investigated the contributions to the CHL reconstruction with changes in the predictors.

Dramatic environmental changes have been recently reported in the Yellow Sea (YS), the South Sea of Korea (SS), and the East/Japan Sea (EJS), but little information on the regional primary productions is currently available. Using the ¹³C-¹⁵N tracer method, we measured primary productions in the YS, the SS, and the EJS for the first time in 2018 to understand the current status of marine ecosystems in the three distinct seas. The mean daily primary productions during the observation period ranged from 25.8 to 607.5 mg C m⁻² d⁻¹ in the YS, 68.5 to 487.3 mg C m⁻² d⁻¹ in the SS, and 106.4 to 490.5 mg C m⁻² d⁻¹ in the EJS, respectively. In comparison with previous studies, significantly lower (t-test, p < 0.05) spring and summer productions and consequently lower annual primary productions were observed in this study. Based on PCA analysis, we found that small-sized (pico- and nano-) phytoplankton had strongly negative effects on the primary productions. Their ecological roles should be further investigated in the YS, the SS, and the EJS under warming ocean conditions within small phytoplankton-dominated ecosystems.

Little information on the phytoplankton community in the Yellow Sea (YS)—especially size-fractionated phytoplankton—is currently available, in comparison to the various physicochemical studies in the literature. Using high-performance liquid chromatography (HPLC), size-fractionated phytoplankton communities were seasonally investigated in the YS in 2019. In the study period, diatoms (55.0 ± 10.2%) and cryptophytes (16.9 ± 9.3%) were the dominant groups. Due to the recent alteration in inorganic nutrient conditions reported in the YS, the contribution of diatoms was lower than in previous studies. The large-sized phytoplankton group (>20 µm) was dominated mostly by diatoms (89.0 ± 10.6%), while the small-sized phytoplankton group (<20 µm) was also dominated by diatoms (41.9 ± 9.1%), followed by cryptophytes (19.2 ± 9.8%). The contributions of small-sized diatoms (<20 µm) have been overlooked in the past, as they are difficult to detect, but this study confirms significant amounts of small-sized diatoms, accounting for 62.3% of the total diatoms in the YS. This study provides an important background for assessing the seasonal variations in different-sized diatom groups in the YS. Further detailed studies on their potential ecological roles should be conducted, in order to better understand marine ecosystems under future warming scenarios.

Transparent exopolymer particles (TEPs) are defined as acidic polysaccharide particles and they are influenced by various biotic and abiotic processes that play significant roles in marine biogeochemical cycles. However, little information on their monthly variation, relationship and contribution to particulate organic carbon (POC) is currently available particularly in coastal regions. In this study, the water samples were collected monthly to determine TEP concentrations and POC concentrations in a southern coastal region of Korea, Jaran Bay from April 2016 to March 2017. The TEP concentrations varied from 26.5 to 1695.4 μg Xeq L⁻¹ (mean ± standard deviation (S.D.) = 215.9 ± 172.2 μg Xeq L⁻¹) and POC concentrations ranged from 109.9 to 1201.9 μg L⁻¹ (mean ± S.D. = 399.1 ± 186.5 μg L⁻¹) during our observation period. Based on the ¹³C stable isotope tracer technique, monthly carbon uptake rates of phytoplankton ranged from 3.0 to 274.1 mg C m⁻² h⁻¹ (mean ± S.D. = 34.5 ± 45.2 mg C m⁻² h⁻¹). The cross-correlation analysis showed a lag-time of 2 months between chlorophyll a and TEP concentrations (r = 0.86, p < 0.01; Pearson’s correlation coefficient). In addition, we observed a 2 month lag-phased correlation between TEP concentrations and primary production (r = 0.73, p < 0.05; Pearson’s correlation coefficient). In Jaran Bay, the TEP contribution was as high as 78.0% of the POC when the TEP-C content was high and declined to 2.4% of the POC when it was low. These results showed that TEP-C could be a significant contributor to the POC pool in Jaran Bay.

Food material (FM) derived from biochemical components (e.g., proteins, lipids, and carbohydrates) of phytoplankton can provide important quantitative and qualitative information of the food available to filter-feeding animals. The main objective of this study was to observe the seasonal and spatial variations of the biochemical compositions of phytoplankton and to identify the major controlling factors of FM as a primary food source in Jaran Bay, a large shellfish aquaculture site in South Korea. Based on monthly sampling conducted during 2016, significant monthly variations in the depth-integrated concentrations of major inorganic nutrients and chlorophyll a within the euphotic water column and a predominance (49.9 ± 18.7%) of micro-sized phytoplankton (>20 μm) were observed in Jaran Bay. Carbohydrates were the dominant biochemical component (51.8 ± 8.7%), followed by lipids (27.3 ± 3.8%) and proteins (20.9 ± 7.4%), during the study period. The biochemical compositions and average monthly FM levels (411.7 ± 93.0 mg m⁻³) in Jaran Bay were not consistent among different bays in the southern coastal region of South Korea, possibly due to differences in controlling factors, such as environmental and biological factors. According to the results from multiple linear regression, the variations in FM could be explained by the relatively large phytoplankton and the P* (PO₄³⁻ − 1/16 × NO₃⁻) and NH₄⁺ concentrations in Jaran Bay. The macromolecular compositions and FM, as alternatives food source materials, should be monitored in Jaran Bay due to recent changes in nutrient concentrations and phytoplankton communities.

The macromolecular concentrations and compositions of phytoplankton are crucial for the growth or nutritional structure of higher trophic levels through the food web in the ecosystem. To understand variations in macromolecular contents of phytoplankton, we investigated the macromolecular components of phytoplankton and analyzed their spatial pattern on the Chukchi Shelf and the Canada Basin. The carbohydrate (CHO) concentrations on the Chukchi Shelf and the Canada Basin were 50.4–480.8 μg L⁻¹ and 35.2–90.1 μg L⁻¹, whereas the lipids (LIP) concentrations were 23.7–330.5 μg L⁻¹ and 11.7–65.6 μg L⁻¹, respectively. The protein (PRT) concentrations were 25.3–258.5 μg L⁻¹ on the Chukchi Shelf and 2.4–35.1 μg L⁻¹ in the Canada Basin. CHO were the predominant macromolecules, accounting for 42.6% on the Chukchi Shelf and 60.5% in the Canada Basin. LIP and PRT contributed to 29.7% and 27.7% of total macromolecular composition on the Chukchi Shelf and 30.8% and 8.7% in the Canada Basin, respectively. Low PRT concentration and composition in the Canada Basin might be a result from the severe nutrient-deficient conditions during phytoplankton growth. The calculated food material concentrations were 307.8 and 98.9 μg L⁻¹, and the average calorie contents of phytoplankton were 1.9 and 0.6 kcal m⁻³ for the Chukchi Shelf and the Canada Basin, respectively, which indicates the phytoplankton on the Chukchi Shelf could provide the large quantity of food material and high calories to the higher trophic levels. Overall, our results highlight that the biochemical compositions of phytoplankton are considerably different in the regions of the Arctic Ocean. More studies on the changes in the biochemical compositions of phytoplankton are still required under future environmental changes.