The adverse health effects associated with the inhalation and ingestion of naturally occurring radon gas produced during the uranium decay chain mean that there is a need to identify high-risk areas. This study detected radon-prone areas using a geographic information system (GIS)-based probabilistic and machine learning methods, including the frequency ratio (FR) model and a convolutional neural network (CNN). Ten influencing factors, namely elevation, slope, the topographic wetness index (TWI), valley depth, fault density, lithology, and the average soil copper (Cu), calcium oxide (Cao), ferric oxide (Fe₂O₃), and lead (Pb) concentrations, were analyzed. In total, 27 rock samples with high activity concentration index values were divided randomly into training and validation datasets (70:30 ratio) to train the models. Areas were categorized as very high, high, moderate, low, and very low radon areas. According to the models, approximately 40% of the study area was classified as very high or high risk. Finally, the radon potential maps were validated using the area under the receiver operating characteristic curve (AUC) analysis. This showed that the CNN algorithm was superior to the FR method; for the former, AUC values of 0.844 and 0.840 were obtained using the training and validation datasets, respectively. However, both algorithms had high predictive power. Slope, lithology, and TWI were the best predictors of radon-affected areas. These results provide new information regarding the spatial distribution of radon, and could inform the development of new residential areas. Radon screening is important to reduce public exposure to high levels of naturally occurring radiation.

South Korea has experienced a rapid increase in ozone concentrations in surface air together with China for decades. Here we use a 3-D global chemical transport model, GEOS-Chem nested over East Asia (110 E - 140 E, 20 N–50 N) at 0.25° × 0.3125° resolution, to examine locally controllable (domestic anthropogenic) versus uncontrollable (background) contributions to ozone air quality at the national scale for 2016. We conducted model simulations for representative months of each season: January, April, July, and October for winter, spring, summer, and fall and performed extensive model evaluation by comparing simulated ozone with observations from satellite and surface networks. The model appears to reproduce observed spatial and temporal ozone variations, showing correlation coefficients (0.40–0.87) against each observation dataset. Seasonal mean ozone concentrations in the model are the highest in spring (39.3 ± 10.3 ppb), followed by summer (38.3 ± 14.4 ppb), fall (31.2 ± 9.8 ppb), and winter (24.5 ± 7.9 ppb), which is consistent with that of surface observations. Background ozone concentrations obtained from a sensitivity model simulation with no domestic anthropogenic emissions show a different seasonal variation in South Korea, showing the highest value in spring (46.9 ± 3.4 ppb) followed by fall (38.2 ± 3.7 ppb), winter (33.0 ± 1.9 ppb), and summer (32.1 ± 6.7 ppb). Except for summer, when the photochemical formation is dominant, the background ozone concentrations are higher than the seasonal ozone concentrations in the model, indicating that the domestic anthropogenic emissions play a role as ozone loss via NOₓ titration throughout the year. Ozone air quality in South Korea is determined mainly by year-round regional background contributions (peak in spring) with summertime domestic ozone formation by increased biogenic VOCs emissions with persistent NOₓ emissions throughout the year. The domestic NOₓ emissions reduce MDA8 ozone around large cities (Seoul and Busan) and hardly increase MDA8 in other regions in spring, but it increases MDA8 across the country in summer. Therefore, NOₓ reduction can be effective in control of MDA8 ozone in summer, but it can have rather countereffect in spring.

Predicting the occurrence of algal blooms is of great importance in managing water quality. Moreover, the demand for predictive models, which are essential tools for understanding the drivers of algal blooms, is increasing with global warming. However, modeling cyanobacteria dynamics is a challenging task. We developed a multivariate Chain-Bernoulli-based prediction model to effectively forecast the monthly sequences of algal blooms considering hydro-environmental predictors (water temperature, total phosphorus, total nitrogen, and water velocity) at a network of stations. The proposed model effectively predicts the risk of harmful algal blooms, according to performance measures based on categorical metrics of a contingency table. More specifically, the model performance assessed by the LOO cross-validation and the skill score for the POD and CSI during the calibration period was over 0.8; FAR and MR were less than 0.15. We also explore the relationship between hydro-environmental predictors and algal blooms (based on cyanobacteria cell count) to understand the dynamics of algal blooms and the relative contribution of each potential predictor. A support vector machine is applied to delineate a plane separating the presence and absence of algal bloom occurrences determined by stochastic simulations using different combinations of predictors. The multivariate Chain-Bernoulli-based prediction model proposed here offers effective, scenario-based, and strategic options and remedies (e.g., controlling the governing environmental predictors) to relieve or reduce increases in cyanobacteria concentration and enable the development of water quality management and planning in river systems.

Contribution of liquid water content (LWC) to the levels of the carcinogenic particulate nitro(so) compounds and the chemistry affecting LWC were investigated based on the observation of seven nitrosamines and two nitramines in rural (Seosan) and urban (Seoul) area in South Korea during October 2019 and a model simulation. The concentrations of both the total nitrosamines and nitramines were higher in Seosan (12.48 ± 16.12 ng/m³ and 0.65 ± 0.71 ng/m³, respectively) than Seoul (7.41 ± 13.59 ng/m³ and 0.24 ± 0.15 ng/m³, respectively). The estimated LWC using a thermodynamic model in Seosan (12.92 ± 9.77 μg/m³) was higher than that in Seoul (6.20 ± 5.35 μg/m³) mainly due to higher relative humidity (75 ± 9% (Seosan); 62 ± 10% (Seoul)) and higher concentrations of free ammonia (0.13 ± 0.09 μmol/m³ (Seosan); 0.08 ± 0.01 μmol/m³ (Seoul)) and total nitric acid (0.09 ± 0.07 μmol/m³ (Seosan); 0.04 ± 0.02 μmol/m³ (Seoul)) in Seosan while neither fog nor rain occurred during the sampling period. The relatively high concentrations of the particulate nitrosamines (>30 ng/m³) only observed probably due to the higher LWC (>10 μg/m³) in Seosan. It implies that aqueous phase reactions involving NO₂ and/or uptake from the gas phase enhanced by LWC could be promoted in Seosan. Strong correlation between the concentrations of nitrosodi-methylamine (NDMA), an example of nitrosamines, simulated by a kinetic box model including the aqueous phase reactions and the measured concentration of NDMA in Seosan (R = 0.77; 0.37 (Seoul)) indicates that the aqueous phase reactions dominantly enhanced the NDMA concentrations in Seosan. On the other hand, it is estimated that the formation of nitrosamines by aqueous phase reaction was not significant due to the relatively lower LWC in Seoul compared to that in Seosan. Furthermore, it is presumed that nitramines are mostly emitted from the primary emission sources. This study implies that the concentration of the particulate nitrosamines can be promoted by aqueous phase reaction enhanced by LWC.

The abuse or misuse of antibiotics is directly linked to the emergence of antibiotic-resistant bacteria and antibiotic resistance genes (ARGs) in the environment. Most fish farms located on Jeju Island operate a flow-through system that pumps in seawater for fish farming and discharges it back to the ocean. To investigate the amount of ARGs that these fish farm effluents discharge into the marine environment, we conducted a metagenomic-based resistome analysis. We observed higher levels of ARGs in fish farm effluents than in seawater at beach and residential areas. A greater proportion of ARGs was found on plasmid rather than on chromosomal DNA, especially for sulfonamide and phenicol classes. The distribution of ARGs did not differ between summer and winter, but the microbial community did. In addition, fish farm samples contained significantly more opportunistic pathogens (i.e., Vibrio, Photobacterium, Aliivibrio, and Tenacibaculum) and virulence factors than non-fish farm samples. Vibrio was the most frequently identified host of ARGs and virulence factors. The presence of Vibrio in the coastal area has been increasing owing to the recent rise in the temperature of seawater. This study suggests the need for actions to treat or monitor ARGs in the coastal areas where fish farms operating a flow-through system are located.

We describe the androgen receptor (AR) agonistic/antagonistic effects of 140 veterinary drugs regulated in Republic of Korea, by setting maximum residue limits. It was conducted using two in vitro test guidelines of the Organization for Economic Cooperation and Development (OECD)—the AR-EcoScreen AR transactivation (TA) assay and the 22Rv1/MMTV_GR-KO AR TA assay. These were performed alongside the AR binding affinity assay to confirm whether their AR agonistic/antagonistic effects are based on the binding affinity to AR. Prior to conducting the AR TA assay, the proficiency test was passed the proficiency performance criterion for the AR agonist and AR antagonist assays. Among the veterinary drugs tested, four veterinary drugs (dexamethasone, trenbolone, altrenogest, and nandrolone) and six veterinary drugs (cymiazole, dexamethasone, zeranol, phenothiazine, bromopropylate, and isoeugenol) were determined as AR agonist and AR antagonist, respectively in both in vitro AR TA assays. Zeranol exhibited weak AR agonistic effects with a PC₁₀ value only in the 22Rv1/MMTV_GR-KO AR TA assay. Regarding changing the AR agonistic/antagonistic effects through metabolism, the AR antagonistic activities of zeranol, phenothiazine, and isoeugenol decreased significantly in the presence of phase I + II enzymes.These data indicate that various veterinary drugs could have the potential to disrupt AR-mediated human endocrine system. Furthermore, this is the first report providing information on AR agonistic/antagonistic effects of veterinary drugs using in vitro OECD AR TA assays.

Per- and polyfluoroalkyl substances (PFAS) concentrations of groundwater in three cities of the Nakdong River Basin in South Korea were quantified to investigate PFAS contamination and the effect of PFAS leakage incident that occurred in the study area in 2018. Groundwater PFASs concentration ranged from non-detectable (N.D.) to 36.9 ng/L (mean 14.1 ng/L), in which, perfluorooctanoic acid (PFOA), perfluorohexanoic acid (PFHxA), perfluoropentanoic acid (PFPeA), and perfluorohexane sulfonate (PFHxS) were commonly observed. Compared to long-chain (C ≥ 8) PFAS, short-chain (<C8) PFAS are more commonly detected in groundwater. Statistical differences were found between the groundwater obtained from different land use. PFAS detected in groundwater from industrial land use were significant different (p<0.01) than other land usages. Spatial difference of PFAS concentrations and distributions in groundwater were also found. PFAS concentrations in groundwater at the furthest downstream area (mean 26.4 ng/L) were the highest followed by the middle reaches (mean 16.2 ng/L), and the upstream area (mean 4.3 ng/L). PFHxS, which was detected dominantly in the middle reach areas, contributed 51% of the total PFAS concentration, but was not detected in the upstream area. There was no health risk by drinking groundwater but found the effect of PFHxS leakage incident on groundwater.

Soil samples were collected at 61 sites of the national monitoring network for persistent organic pollutants (POPs) in South Korea. The target compounds were brominated flame retardants (BFRs), including polybrominated diphenyl ethers (PBDEs), polybrominated biphenyls (PBBs), hexabromocyclododecanes (HBCDDs), and tetrabromobisphenol A (TBBPA). The mean concentrations of Σ₂₇ PBDEs, Σ₃ HBCDDs, and TBBPA in soil were 222, 17.2, and 4.4 ng/g, respectively, but PBBs were not detected. Industrial sites had statistically higher BFR concentrations than suburban sites but no significant difference compared with urban sites. The commercial deca-BDE mixtures were the most likely source of PBDE contamination in the soil samples, with the minor influence of commercial penta-BDE and octa-BDE mixtures. The profiles of HBCDDs in most soil samples differed from those in the powder types of technical HBCDD mixtures, indicating that they are affected by the HBCDDs contained in commercial products and the conversion of HBCDD diastereoisomers (γ-HBCDD to α-HBCDD) in the environment. The concentrations of Σ₂₇ PBDEs, Σ₃ HBCDDs, and TBBPA were significantly correlated with population density, gross domestic product, and the number of companies (p < 0.01), indicating a direct impact of anthropogenic activities. Significant correlations among BFRs were determined (0.63 < r < 0.74, p < 0.01), suggesting that these pollutants had similar sources. Relatively good correlations (0.44 < r < 0.98, p < 0.01) between BDE-209 and other light BDEs (except for BDE-71, -77, −126, −156, and −205) might result from the degradation of heavy BDEs under anaerobic and natural sunlight conditions. To the best of our knowledge, this study provides the most comprehensive soil monitoring data for various BFRs in South Korea. Furthermore, it is the first report on soil contamination by deca-BDE, HBCDDs, and TBBPA in South Korea.

The Hebei Spirit oil spill (HSOS) occurred on the west coast of South Korea (Taean county) on December 7, 2007, and studies revealed that exposure to the oil spill was associated with various adverse health issues in the inhabiting population. However, no studies evaluated the association between crude-oil exposure and epigenetic changes. This study aimed to investigate the HSOS exposure-associated longitudinal and cross-sectional variations in global DNA methylation (5-mc) and/or hydroxymethylation (5-hmc) and expression profiles of related genes in Taean cohort participants from 2009 (AH-baseline) and 2014 (AH-follow-up) relative to the reference group (AL). We measured global DNA 5-mc and 5-hmc levels and related gene expression levels in whole blood. We identified significant associations between HSOS exposure and AH-baseline-5-mc, AH-baseline-5-hmc, and AH-follow-up-5-hmc. HSOS exposure was associated with lower %5-mc content and higher %5-hmc content in the same individuals from both the cross-sectional and longitudinal studies. In addition, we found a strong correlation between 5-mc and DNMT3B expression, and between 5-hmc and TET1 expression. Our findings suggested that epigenetic changes are important biomarkers for HSOS exposure and that 5-hmc is likely to be more sensitive for environmental epidemiological studies.

Trophic magnification factor (TMF) of persistent toxic substances (PTSs: Hg, PCBs, PAHs, and styrene oligomers (SOs)) in a coastal food web (12 fish and four invertebrates) was determined in Ulsan Bay, South Korea. The nitrogen stable isotope ratios (δ¹⁵N) of amino acids [δ¹⁵NGₗᵤ₋Pₕₑ based on glutamic acid (δ¹⁵NGₗᵤ) and phenylalanine (δ¹⁵NPₕₑ)] were used to estimate the trophic position (TPGₗᵤ₋Pₕₑ) of organisms. The TPGₗᵤ₋Pₕₑ of organisms ranged from 1.64 to 3.69, which was lower than TP estimated by δ¹⁵N of bulk particulate organic matter (TPBᵤₗₖ: 2.46–4.21). Mercury and CB 138, 153, 187, and 180 were biomagnified through the whole food web (TMF > 1), while other PTSs, such as PAHs and SOs were not (biodilution of SOs firstly reported). In particular, the trophic transfer of PTSs was pronounced in the resident fish (e.g., rock bream, sea perch, Korean rockfish). Of note, CB 99, 101, 118, and 183 were additionally found to be biomagnifying PTSs in these species. Thus, fish residency appears to represent an important factor in determining the TMF of PTSs in the coastal environment. Overall, δ¹⁵NGₗᵤ₋Pₕₑ provided accurate TPs of organisms and could be applied to determine the trophic transfer of PTSs in coastal food webs.