Machine Learning-Based Early Warning Level Prediction for Cyanobacterial Blooms Using Environmental Variable Selection and Data Resampling.

Many countries have attempted to mitigate and manage issues related to harmful algal blooms (HABs) by monitoring and predicting their occurrence. The infrequency and duration of HABs occurrence pose the challenge of data imbalance when constructing machine learning models for their prediction. Furthermore, the appropriate selection of input variables is a significant issue because of the complexities between the input and output variables. Therefore, the objective of this study was to improve the predictive performance of HABs using feature selection and data resampling. Data resampling was used to address the imbalance in the minority class data. Two machine learning models were constructed to predict algal alert levels using 10 years of meteorological, hydrodynamic, and water quality data. The improvement in model accuracy due to changes in resampling methods was more noticeable than the improvement in model accuracy due to changes in feature selection methods. Models constructed using combinations of original and synthetic data across all resampling methods demonstrated higher prediction performance for the caution level (L-1) and warning level (L-2) than models constructed using the original data. In particular, the optimal artificial neural network and random forest models constructed using combinations of original and synthetic data showed significantly improved prediction accuracy for L-1 and L-2, representing the transition from normal to bloom formation states in the training and testing steps. The test results of the optimal RF model using the original data indicated prediction accuracies of 98.8% for L0, 50.0% for L1, and 50.0% for L2. In contrast, the optimal random forest model using the Synthetic Minority Oversampling Technique-Edited Nearest Neighbor (ENN) sampling method achieved accuracies of 85.0% for L0, 85.7% for L1, and 100% for L2. Therefore, applying synthetic data can address the imbalance in the observed data and improve the detection performance of machine learning models. Reliable predictions using improved models can support the design of management practices to mitigate HABs in reservoirs and ultimately ensure safe and clean water resources.

Molecular-level chemical composition of aerosol and its potential source tracking at Antarctic Peninsula.

Marine organic aerosols play crucial roles in global climatic systems. However, their chemical properties and relationships with various potential organic sources still need clarification. This study employed high-resolution mass spectrometry to investigate the identity, origin, and transportation of organic aerosols in pristine Antarctic environments (King Sejong Station; 62.2°S, 58.8°W), where complex ocean-cryosphere-atmosphere interactions occur. First, we classified the aerosol samples into three clusters based on their air mass transport history. Next, we investigated the relationship between organic aerosols and their potential sources, including organic matter dissolved in the open ocean, coastal waters, and runoff waters. Cluster 1 (C1), in which the aerosols mainly originated from the open ocean area (i.e., pelagic zone-influenced), exhibited a higher abundance of lipid-like and protein-like organic aerosols than cluster 3 (C3), with ratios 1.8- and 1.6-times higher, respectively. In contrast, C3, characterized by longer air mass retention over sea ice and land areas (i.e., inshore-influenced), had higher lignin- and condensed aromatic structures (CAS)-like organic aerosols by 2.2- and 3.4-times compared to C1. Cluster 2 (C2) has intermediate characteristics between C1 and C3 concerning the chemical properties of the aerosols and air mass travel history. Notably, the chemical properties of the aerosols assigned to C1 are closely related to those of phytoplankton-derived organics enriched in the open ocean. In contrast, those of C3 are comparable to those of terrestrial plant-derived organics enriched in coastal and runoff waters. These findings help evaluate the source-dependent properties of organic aerosols in changing Antarctic environment.

Ocean-atmosphere interactions: Different organic components across Pacific and Southern Oceans.

Sea spray aerosol (SSA) particles strongly influence clouds and climate but the potential impact of ocean microbiota on SSA fluxes is still a matter of active research. Here-by means of in situ ship-borne measurements-we explore simultaneously molecular-level chemical properties of organic matter (OM) in oceans, sea ice, and the ambient PM2.5 aerosols along a transect of 15,000 km from the western Pacific Ocean (36°13'N) to the Southern Ocean (75°15'S). By means of orbitrap mass spectrometry and optical characteristics, lignin-like material (24 ± 5 %) and humic material (57 ± 8 %) were found to dominate the pelagic Pacific Ocean surface, while intermediate conditions were observed in the Pacific-Southern Ocean waters. In the marine atmosphere, we found a gradient of features in the aerosol: lignin-like material (31 ± 9 %) dominating coastal areas and the pelagic Pacific Ocean, whereas lipid-like (23 ± 16 %) and protein-like (11 ± 10 %) OM controlled the sympagic Southern Ocean (sea ice-influence). The results of this study showed that the OM composition in the ocean, which changes with latitude, affects the OM in aerosol compositions in the atmosphere. This study highlights the importance of the global-scale OM monitoring of the close interaction between the ocean, sea ice, and the atmosphere. Sympagic primary marine aerosols in polar regions must be treated differently from other pelagic-type oceans.

A novel method for micropollutant quantification using deep learning and multi-objective optimization.

Micropollutants (MPs) released into aquatic ecosystems have adverse effects on public health. Hence, monitoring and managing MPs in aquatic systems are imperative. MPs can be quantified by high-resolution mass spectrometry (HRMS) with stable isotope-labeled (SIL) standards. However, high cost of SIL solutions is a significant issue. This study aims to develop a rapid and cost-effective analytical approach to estimate MP concentrations in aquatic systems based on deep learning (DL) and multi-objective optimization. We hypothesized that internal standards could quantify the MP concentrations other than the target substance. Our approach considered the precision of intra-/inter-day repeatability and natural organic matter information to reduce instrumental error and matrix effect. We selected standard solutions to estimate the concentrations of 18 MPs. Among the optimal DL models, DarkNet-53 using nine standard solutions yielded the highest performance, while ResNet-50 yielded the lowest. Overall, this study demonstrated the capability of DL models for estimating MP concentrations.

Abundance and diversity of antibiotic resistance genes and bacterial communities in the western Pacific and Southern Oceans.

This study investigated the abundance and diversity of antibiotic resistance genes (ARGs) and the composition of bacterial communities along a transect covering the western Pacific Ocean (36°N) to the Southern Ocean (74°S) using the Korean icebreaker R/V Araon (total cruise distance: 14,942 km). The relative abundances of ARGs and bacteria were assessed with quantitative PCR and next generation sequencing, respectively. The absolute abundance of ARGs was 3.0 × 106 ± 1.6 × 106 copies/mL in the western Pacific Ocean, with the highest value (7.8 × 106 copies/mL) recorded at a station in the Tasman Sea (37°S). The absolute abundance of ARGs in the Southern Ocean was 1.8-fold lower than that in the western Pacific Ocean, and slightly increased (0.7-fold) toward Terra Nova Bay in Antarctica, possibly resulting from natural terrestrial sources or human activity. ß-Lactam and tetracycline resistance genes were dominant in all samples (88-99%), indicating that they are likely the key ARGs in the ocean. Correlation and network analysis showed that Bdellovibrionota, Bacteroidetes, Cyanobacteria, Margulisbacteria, and Proteobacteria were positively correlated with ARGs, suggesting that these bacteria are the most likely ARG carriers. This study highlights the latitudinal profile of ARG distribution in the open ocean system and provides insights that will help in monitoring emerging pollutants on a global scale.

When river water meets seawater: Insights into primary marine aerosol production.

The impact of inorganic salts and organic matter (OM) on the production of primary marine aerosols is still under debate. To constrain their impact, we investigated primary aerosols generated by a sea-spray generator chamber using surface water samples from rivers, estuaries, and seas that were collected along salinity gradients in two temperate Korean coastal systems and one Arctic coastal system. Salinity values showed an increasing trend along the river-estuary-coastal water transition, indicating the lowest amount of inorganic salts in the river but the highest amount in the sea. In river samples, the lowest number concentration of primary aerosol particles (1.01 × 103 cm-3) was observed at the highest OM content, suggesting that low salinity controls aerosol production. Moreover, the number concentration of primary aerosols increased drastically in estuarine (1.13 × 104 cm-3) and seawater (1.35 × 104 cm-3) samples as the OM content decreased. Our results indicate that inorganic salts associated with increasing salinity play a much larger role than OM in aerosol production in river-dominated coastal systems. Laboratory studies using NaCl solution supported the conclusion that inorganic salt is a critical factor in modulating the particles produced from river water and seawater. Accordingly, this study highlights that inorganic salts are a critical factor in modulating the production of primary marine aerosols.

Prediction of antibiotic-resistance genes occurrence at a recreational beach with deep learning models.

Antibiotic resistance genes (ARGs) have been reported to threaten the public health of beachgoers worldwide. Although ARG monitoring and beach guidelines are necessary, substantial efforts are required for ARG sampling and analysis. Accordingly, in this study, we predicted ARGs occurrence that are primarily found on the coast after rainfall using a conventional long short-term memory (LSTM), LSTM-convolutional neural network (CNN) hybrid model, and input attention (IA)-LSTM. To develop the models, 10 categories of environmental data collected at 30-min intervals and concentration data of 4 types of major ARGs (i.e., aac(6'-Ib-cr), blaTEM, sul1, and tetX) obtained at the Gwangalli Beach in South Korea, between 2018 and 2019 were used. When individually predicting ARGs occurrence, the conventional LSTM and IA-LSTM exhibited poor R2 values during training and testing. In contrast, the LSTM-CNN exhibited a 2-6-times improvement in accuracy over those of the conventional LSTM and IA-LSTM. However, when predicting all ARGs occurrence simultaneously, the IA-LSTM model exhibited a superior performance overall compared to that of LSTM-CNN. Additionally, the influence of environmental variables on prediction was investigated using the IA-LSTM model, and the ranges of input variables that affect each ARG were identified. Consequently, this study demonstrated the possibility of predicting the occurrence and distribution of major ARGs at the beach based on various environmental variables, and the results are expected to contribute to management of ARG occurrence at a recreational beach.

Hydrometeorological Influence on Antibiotic-Resistance Genes (ARGs) and Bacterial Community at a Recreational Beach in Korea.

We investigated the occurrence and distribution of antibiotic-resistance genes (ARGs) and the composition of a bacterial community under conditions of rainfall on a recreational beach in Korea. Seawater samples, collected every 1‒5 hours in June 2018 and May 2019, were analyzed using quantitative real-time polymerase chain reaction and next-generation sequencing. We found a substantial influence of rainfall and tidal levels on the relative abundance of total ARGs and bacterial operational taxonomic units (OTUs), which showed 1.9 × 103 and 1.1 × 101 fold increases, respectively. In particular, the elevated levels of ARGs were maintained for up to 32 hours after rainfall. An increased abundance of sewage-related ARGs and bacterial OTUs suggested that combined sewer overflow (CSO) may be the major factor contributing to the increase in the number and diversity of ARGs and related bacterial communities. Network analysis of ARGs and OTUs indicated that, at the genus level, Acinetobacter, Pseudomonas, and Prevotella were the main potential pathogens carrying the observed ARGs in the recreational seawater. Overall, these findings highlight the potential threat to public health on beaches, and indicate the requirement for more adequate monitoring, with greater efforts to mitigate the propagation of ARGs arising from CSOs.

Single- and mixture toxicity of three organic UV-filters, ethylhexyl methoxycinnamate, octocrylene, and avobenzone on Daphnia magna.

In freshwater environments, aquatic organisms are generally exposed to mixtures of various chemical substances. In this study, we tested the toxicity of three organic UV-filters (ethylhexyl methoxycinnamate, octocrylene, and avobenzone) to Daphnia magna in order to evaluate the combined toxicity of these substances when in they occur in a mixture. The values of effective concentrations (ECx) for each UV-filter were calculated by concentration-response curves; concentration-combinations of three different UV-filters in a mixture were determined by the fraction of components based on EC25 values predicted by concentration addition (CA) model. The interaction between the UV-filters were also assessed by model deviation ratio (MDR) using observed and predicted toxicity values obtained from mixture-exposure tests and CA model. The results from this study indicated that observed ECxmix (e.g., EC10mix, EC25mix, or EC50mix) values obtained from mixture-exposure tests were higher than predicted ECxmix (e.g., EC10mix, EC25mix, or EC50mix) values calculated by CA model. MDR values were also less than a factor of 1.0 in a mixtures of three different UV-filters. Based on these results, we suggest for the first time a reduction of toxic effects in the mixtures of three UV-filters, caused by antagonistic action of the components. Our findings from this study will provide important information for hazard or risk assessment of organic UV-filters, when they existed together in the aquatic environment. To better understand the mixture toxicity and the interaction of components in a mixture, further studies for various combinations of mixture components are also required.

Morphological, molecular, and biochemical characterization of astaxanthin-producing green microalga Haematococcus sp. KORDI03 (Haematococcaceae, Chlorophyta) isolated from Korea.

A unicellular red microalga was isolated from environmental freshwater in Korea, and its morphological, molecular, and biochemical properties were characterized. Morphological analysis revealed that the isolate was a unicellular biflagellated green microalga that formed a non-motile, thick-walled palmelloid or red aplanospore. To determine the taxonomical position of the isolate, its 18S rRNA and rbcL genes were sequenced and phylogenetic analysis was performed. We found that the isolate was clustered together with other related Haematococcus strains showing differences in the rbcL gene. Therefore, the isolated microalga was classified into the genus Haematococcus, and finally designated Haematococcus sp. KORDI03. The microalga could be cultivated in various culture media under a broad range of pH and temperature conditions. Compositions of the microalgal cellular components were analyzed, and its protein, carbohydrate, and lipid compositions were estimated to be 21.1 ± 0.2%, 48.8 ± 1.8%, and 22.2 ± 0.9%, respectively. In addition, D-glucose and D-mannose were the dominant monosaccharides in the isolate, and its amino acids were composed mainly of aspartic acid, glutamic acid, alanine, and leucine. Moreover, several polyunsaturated fatty acids accounted for about 80% of the total fatty acids in Haematococcus sp. KORDI03, and the astaxanthin content in the red aplanospores was estimated to be 1.8% of the dry cell weight. To the best of our knowledge, this is the first report of an Haematococcus sp. isolated from Korea, which may be used for bioresource production in the microalgal industry.

Chromene suppresses the activation of inflammatory mediators in lipopolysaccharide-stimulated RAW 264.7 cells.

Inflammation is complex process involving a variety of immune cells that defend the body from harmful stimuli. However, pro-inflammatory cytokines and inflammatory mediators can also exacerbate diseases such as cancer. The aim of this study was to identify a natural effective remedy for inflammation. We isolated a functional algal chromene compound from Sargassum siliquastrum, named sargachromanol D (SD). We evaluated the anti-inflammatory effect of SD on lipopolysaccharide (LPS)-exposed RAW 264.7 cells by measuring cell viability, cytotoxicity, and production of inflammatory mediators such as nitric oxide (NO), prostaglandin E2 (PGE2), inducible NO synthase (iNOS), cyclooxygenase-2 (COX-2), and the pro-inflammatory cytokines tumor necrosis factor-α (TNF-α), interleukin (IL)-1ß, and IL-6. SD inhibited production of NO and PGE2 from LPS-induced cells by preventing the expression of inflammatory mediators such as iNOS and COX-2 in a dose-dependent manner. Concurrently, levels of the pro-inflammatory cytokines TNF-α, IL-1ß, and IL-6 were reduced with increasing concentrations of SD. In addition, SD inhibited the activation of NF-κB and mitogen-activated protein kinases (MAPKs) pathways in a concentration-dependent manner. These results indicate that SD inhibits LPS-stimulated inflammation by inhibition of the NF-κB and MAPKs pathways in macrophages.

Influence of microorganism content in suspended particles on the particle-water partitioning of mercury in semi-enclosed coastal waters.

It is known that particle scavenging of mercury (Hg) can be affected by the abundance of particulate organic matter in coastal waters. However, the role of living organic particles in Hg scavenging is not yet completely understood. In this study, we hypothesized that an abundance of living organic particles (i.e., phytoplankton and bacteria) would influence the particle-water partitioning of Hg in coastal waters. Surface seawater samples were collected from eight stations in Gwangyang Bay, Korea, in three seasons (November 2009, April 2010, and October 2010) for the determination of concentrations of suspended particulate matter (including chlorophyll-a and bacteria), and Hg in unfiltered and filtered waters. We found that more Hg partitioned toward particulate matter when phytoplankton biomass, indicated from the chlorophyll-a concentration in a particle, was higher. In the low algal season, when [chlorophyll-a]<0.6 µg L(-1), the bacterial number, instead of chlorophyll-a concentration in particle, showed a positive correlation with the particle-water partition coefficient of Hg. Overall, microbial abundance seems to play a critical role in particle scavenging of Hg in coastal water. Taking this result in light of Hg in pristine coastal zones, we predict that increases in algal biomass amplify the potential for algae to transfer Hg to marine food chains.

Algivirga pacifica gen. nov., sp. nov., a novel agar-degrading marine bacterium of the family Flammeovirgaceae isolated from Micronesia.

An aerobic, Gram-negative, coccoid to short rod-shaped and non-flagellated marine bacterial strain S354(T) was isolated from seawater of Micronesia. The strain was capable to degrade agar-forming slight depression into agar plate. Growth occurred at a temperature range of 12-44 °C, a pH range of 5-9, and a salinity range of 1-7 % (w/v) NaCl. Phylogenetic analyses based on 16S rRNA gene sequences suggested that S354(T) belongs to the family Flammeovirgaceae. The novel strain was most closely related to Limibacter armeniacum YM 11-185(T) with similarity of 92.5 %. The DNA G+C content was 43.8 mol%. The major fatty acids (>10 %) were iso-C15:0 and C16:1 ω5c. The predominant isoprenoid quinone was determined to be MK-7. Polar lipid profile of S354(T) consisted of phosphatidylethanolamine, unknown polar lipid, and unknown glycolipids. Based on the phenotypic, phylogenetic, biochemical, and physiological tests conducted in this study, S354(T) is proposed to represent a type strain of a novel genus and species. The 16S rRNA gene sequence of S354(T) is registered in GenBank under the accession number JQ639084. The type of strain Algivirga pacifica gen. nov., sp. nov. is S354(T) (=KCCM 90107(T)=JCM 18326(T)).

Draft genome sequence of the xylan-degrading marine bacterium strain S124, representing a novel species of the genus Oceanicola.

We isolated a xylan-degrading bacterium from seawater of Micronesia and identified it as Oceanicola sp. strain S124. We sequenced the Oceanicola sp. S124 genome using GSFLX 454 pyrosequencing and predicted 4,433 open reading frames (ORFs) including putative saccharification and phage-related genes.

Photo-oxidative stress by ultraviolet-B radiation and antioxidative defense of eckstolonol in human keratinocytes.

Ultraviolet-B (UV-B) irradiation has been known to generate oxidative stress by increasing reactive oxygen species (ROS) in skin cells. Several naturally occurring antioxidant compounds isolated from marine algae are believed to protect against ROS. In this study, we assessed the antioxidative effect of eckstolonol isolated from Ecklonia cava against UV-B-induced ROS in human keratinocytes (HaCaTs). We investigated the effects of photo-oxidative stress by UV-B (50 mJ/cm(2)) and the antioxidative effects of eckstolonol using fluorometry, flow cytometry, microscopy, and cell viability and comet assays. UV-B irradiation decreased cell viability, which was restored in a dose-dependent manner with eckstolonol treatment (0, 5, 50, 100, and 200 µM). Moreover, eckstolonol reduced UV-B-induced ROS, lipid peroxidation, damaged DNA levels, and cell death. These antioxidative effects seem to be due to the enzymatic activities of catalase (CAT) and superoxide dismutase (SOD). Collectively, these results indicate that eckstolonol is capable of protecting keratinocytes from photo-oxidative stress.

Importance of monsoon rainfall in mass fluxes of filtered and unfiltered mercury in Gwangyang Bay, Korea.

We investigated the effects of the East Asian Summer Monsoon (EASM), which brings approximately half of Korea's annual rainfall in July, on the concentration and particle-water partitioning, and sources of Hg in coastal waters. Surface seawater samples were collected from eight sites in Gwangyang Bay, Korea, during the monsoon (July, 2009) and non-monsoon dry (April and November, 2009) seasons and the concentrations of suspended particulate matter, chlorophyll-a, and unfiltered and filtered Hg were determined. We found significant (p<0.05) increases in filtered Hg in the monsoon season (1.8 ± 0.019 pM) compared to the dry season (0.62 ± 0.047 pM). In contrast, the Hg concentrations associated with particles showed no significant differences (p>0.05) between the monsoon (459 ± 141 pmol g(-1)) and the dry season (346 ± 30 pmol g(-1)), which resulted in decreased particle-water partition coefficients of Hg in the monsoon season compared to the values in the dry season: 5.7 ± 0.1 in April, 5.3 ± 0.1 in July, and 5.8 ± 0.1 in November. The annual Hg input to Gwangyang Bay was estimated at 64 ± 6.6 mol yr(-1) and 27 ± 1.9 mol yr(-1) for unfiltered and filtered Hg, respectively. The Hg discharged from rivers was a major source of Hg in Gwangyang Bay: the river input contributed 83 ± 13% of total input of unfiltered and 73 ± 6.0% of filtered Hg. On a monthly basis, unfiltered Hg input was 17 ± 11 mol month(-1) in the monsoon season and 3.2 ± 0.70 mol month(-1) in the dry season, while filtered Hg input was 7.1 ± 4.1 mol month(-1) in the monsoon and 1.3 ± 0.26 mol month(-1) in the dry. Consequently, the EASM resulted in an unfiltered Hg input 5.3 times greater than the mean dry month input and a filtered Hg input 5.5 times greater than the mean dry month input, which is mainly attributable to enhanced river water discharge during the monsoon season.