https://doi.org/10.15681/KSWE.2026.42.3.207

정찬영(Chanyoung Jeong) ; 송재하(Jeaha Song) ; 이수빈(Soobeen Lee) ; 최형식(Hyeongsik Choi) ; 김아름(Areum Kim) ; 추연수(Yeounsu Chu) ; 공동수(Dongsoo Kong)

This study evaluates climate change vulnerability in Bangtaecheon, South Korea, by integrating 13 years of September meteorological data with field measurements of water temperature and quality, as well as quantitative benthic macroinvertebrate surveys conducted in September 2012 and twice in September 2025 at six sites along an altitudinal gradient. September air temperature and precipitation trends were analyzed using the Mann?Kendall test with Sen's slope, while community-environment relationships were assessed through redundancy analysis (RDA) incorporating water temperature, electrical conductivity (EC), dissolved oxygen, and pH. Over the last decade, September air temperature and rainfall have increased significantly, whereas mean September stream temperature has not shown a significant monotonic trend. However, a short-term warming event in early September 2025 coincided with notable community reorganization: climate-sensitive taxa declined in middle and downstream reaches, while tolerant groups increased. KTI-based indicator taxa and EPT richness became increasingly restricted to upstream cold-water reaches with lower temperatures and EC, indicating an upslope contraction of suitable habitat. The RDA revealed that temperature, EC, dissolved oxygen, and pH accounted for 31.5% of community variation, highlighting distinct spatial and temporal gradients along the stream continuum. These findings support a zone-based management approach that prioritizes the protection of upstream cold-water refugia, strengthens early-warning monitoring in mid-reaches, and implements riparian buffering and sediment control actions in downstream transition reaches to sustain mountain headwaters amid increasing climate extremes.

https://doi.org/10.15681/KSWE.2026.42.3.220

문예찬(Yechan Moon)

Climate change is increasing extreme precipitation and flood risks in transboundary river basins where sharing hydrological information is limited. The Imjin River basin on the Korean Peninsula illustrates this challenge, as a significant portion of its upstream catchment is in North Korea, where access to in situ observations and dam operation data is severely restricted. In this context, upstream reservoirs like Hwanggang Dam have been identified as potential sources of downstream flood hazards, yet systematic risk assessment remains challenging. This study introduces a satellite-based analytical framework to detect hydrological variability and assess potential dam release risks around Hwanggang Dam using long-term Sentinel-1 Synthetic Aperture Radar (SAR) observations. The analysis combines time-series decomposition, anomaly detection, and predictive modeling to identify both recurring seasonal patterns and deviations linked to increased hydrological risk. To enhance robustness, statistical time-series models are integrated with machine-learning techniques through an ensemble forecasting strategy that considers both long-term structural trends and short-term fluctuations. The results indicate significant seasonal regularity and non-linear trend changes in hydrological conditions near Hwanggang Dam, suggesting that flood risk is influenced by the interplay of climatic seasonality and longer-term structural factors rather than random variability alone. This study highlights the potential of satellite SAR-based time-series analysis as a non-intrusive early warning tool for flood risk management in transboundary river basins affected by climate change.

https://doi.org/10.15681/KSWE.2026.42.3.229

최수빈(Subin Choi) ; 서유진(Yujin Seo) ; 류혁진(Hyeoukjin Ryu) ; 이재건(Jaegeon Lee) ; 심중표(Joongpyo Shim) ; 한협조(Hyeop-Jo Han) ; 조민(Min Cho) ; 윤영건(Younggun Yoon) ; 신재돈(Jaedon Shin)

Sulfate-rich industrial wastewaters generated during secondary battery manufacturing (e.g., Li-ion production and hydrometallurgical recycling) present a growing environmental challenge, necessitating low-energy treatment approaches capable of functioning under high ionic strength. This study assesses a marine microbial consortium enriched from the Saemangeum tidal flat for its ability to couple anaerobic respiration with iron-sulfur (Fe-S) mineralization. This process transforms amorphous Fe(III) minerals and sulfate into biogenic iron sulfide nanominerals. Under ambient batch incubation conditions (30 °C, near-neutral pH) with pyruvate as the electron donor, ferrihydrite acted as a dynamic electron sink, facilitating the coupled reduction of Fe(III) and sulfate. This process resulted in the rapid precipitation of mackinawite-like FeS and FeS2 nanoparticles, which exhibited rod-like and framboid-like morphologies. Time-resolved kinetics indicated that pyruvate oxidation to acetate supported over 60% sulfate reduction across three representative consortia (P1-5, P1-18, P2-30) selected from a total of 105 tested communities. Mineralogical analyses (XRD, TEM, SEM, EDS) revealed a sequential transformation from <5 nm amorphous ferrihydrite to semi-crystalline α-/γ-type Fe(III) oxyhydroxides, followed by the formation of crystalline Fe-S phases. Representative isolates from the P1-5 consortium, recovered on pyruvate- and sulfate-supplemented agar, were identified as Vibrio spp. These findings underscore the potential of halotolerant marine consortia as a viable, low-energy bioprocess for treating sulfate-laden industrial brines and for advancing biologically driven Fe-S mineralization technologies.

https://doi.org/10.15681/KSWE.2026.42.3.243

황인태(Hwang In-tae) ; 김양완(Kim Yang-wan) ; 김기영(Kim Ki-young) ; 박종민(Park Jong-min)

Recent climate change has led to increased water temperatures and altered precipitation patterns, which have intensified harmful algal blooms in major reservoirs across Korea. This situation underscores the necessity for continuous and extensive spatial monitoring. In response, this study developed a machine learning framework to estimate Total Phosphorus (TP) and Chlorophyll-a (Chl-a) concentrations in Daecheong Lake, utilizing Harmonized Landsat Sentinel-2 (HLS) satellite imagery and the eXtreme Gradient Boosting (XGBoost) model. Following this, trophic conditions were assessed using the Trophic State Index (TSI). HLS surface reflectance data from 2013 to 2024 were combined with in situ measurements from national monitoring networks. SHAP (Shapley Additive Explanations) analysis was conducted to identify the most effective spectral inputs, revealing that the Aerosol (0.43?0.45 μm), Green (0.53?0.59 μm), and NIR (Band 8A) bands were the most significant contributors to predictions of both TP and Chl-a. Model validation showed correlation coefficients (R) of approximately 0.62 for both variables, with RMSE and MAE values of 0.018 mg/L and 0.010 mg/L for TP, and 0.023 mg/L and 0.011 mg/L for Chl-a, respectively. While the model successfully captured overall temporal trends, it tended to underestimate peak concentrations during summer months. The spatial distributions of TSI, derived from predicted TP and Chl-a, closely aligned with periods of official algal bloom warnings, with Chl-a-based TSI exhibiting greater sensitivity to short-term environmental changes. These findings demonstrate the effectiveness of HLS-based XGBoost modeling for monitoring eutrophication in large reservoirs.

https://doi.org/10.15681/KSWE.2026.42.3.254

공동수(Dongsoo Kong) ; 송재하(Jea-ha Song) ; 정찬영(Chanyoung Jeong) ; 김명철(Myoung Chul Kim) ; 이종현(Jong Hyeon Lee) ; 곽인실(Ihn-Sil Kwak)

This study identified 25 species of oligochaetes from fine-grained sediments collected across 320 sampling units at 80 sites in rivers and lakes within Korea's five major river systems. The ecological niches of each species were analyzed in relation to environmental factors, including sediment organic matter and heavy metal concentrations. Limnodrilus hoffmeisteri and Tubifex tubifex were the most frequently observed and numerically abundant species, indicating their widespread distribution in freshwater fine sediments in Korea. Canonical correspondence analysis (CCA) revealed that the first axis was strongly correlated with all measured environmental variables, particularly total ammonia nitrogen (TAN) and mud content, suggesting a nutrient and organic matter gradient. The second axis was associated with acid volatile sulfide (AVS), mud content, and heavy metals, reflecting redox conditions and levels of contamination. The third axis correlated with total organic carbon (TOC), indicating an organic matter gradient. Species composition exhibited patterns consistent with these environmental gradients. Tolerance analysis utilizing a Weibull model showed that L. hoffmeisteri and T. tubifex had broad tolerance ranges. Both L. hoffmeisteri and Henlea sp. demonstrated high tolerance to mud, AVS, and heavy metals. Conversely, Lumbriculus variegatus and Nais variabilis exhibited low tolerance to mud and heavy metals but high tolerance to TOC and TAN. Eisenia koreana was sensitive to all environmental variables, while Bothrioneurum vejdovskyanum displayed overall weak tolerance. Additionally, Haplotaxis gordioides and Branchiura sowerbyi were sensitive to TAN.

https://doi.org/10.15681/KSWE.2026.42.3.267

강덕준(Dejun Jiang) ; 권혁구(Hyuk-Ku Kwon)

Predicting chlorophyll-a (Chl-a) concentrations in riverine systems remains a significant challenge for water quality management. The Gapcheon River basin exemplifies this issue, as conventional regression models consistently underestimate peak events due to the statistical rarity of high-concentration observations. This study introduces an end-to-end deep learning framework that addresses data quality and distributional bias without generating synthetic samples. It utilizes 7-day input sequences to predict the next-step Chl-a concentration. To enhance accuracy, density-based anomaly detection using the local outlier factor (LOF) was employed to selectively eliminate non-representative records while preserving valid seasonal extremes. A sigmoid-weighted cost-sensitive loss function was then introduced to focus the model's attention on the underrepresented bloom tail. This allowed bidirectional gated recurrent unit (GRU) and long short-term memory (LSTM) networks to prioritize peak magnitude and the dynamics of rising limbs. A systematic evaluation across three weighting regimes and five random seeds demonstrated that GRU consistently outperformed LSTM. The moderate Thresh_0.75 setting yielded the best mean performance in the high-concentration subset, with the GRU Baseline achieving an overall R² of 0.87 ± 0.01 and RMSE of 3.84 ± 0.14 μg/L, while GRU Thresh_0.75 reached a high-concentration R² of 0.53 ± 0.03 and RMSE of 6.37 ± 0.19 μg/L. Permutation-based feature attribution analysis indicated that cost-sensitive training relied heavily on hydrological dilution signals, as well as dissolved oxygen and pH levels, which are indicators of acute eutrophication. These findings demonstrate that targeted loss modification provides a coherent and plausible approach for forecasting high concentrations in continuously monitored river systems.