Bongseok Jeong

Mr. Bongseok Jeong

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 June 2025 in Water
Verified
Journal Article
Open access
PDF file
BJBongseok Jeong
Bongseok Jeong
SLSunmin Lee
Sunmin Lee
Joonghyeok HeoJoonghyeok Heo
Joonghyeok Heo
JLJeongho Lee
Jeongho Lee
Moung-Jin LeeMoung-Jin Lee
Moung-Jin Lee
Remote sensing and AI models have been utilized for monitoring Chlorophyll-a (Chl-a), a primary indicator of eutrophication across broad water bodies. Previous studies have primarily relied on optical remote sensing data for assessing Chl-a’s spectral characteristics. Synthetic-aperture radar (SAR) data, which contain valuable information about surface algae containing Chl-a, remains underutilized despite its high potential for improving Chl-a retrieval accuracy. Therefore, this study aims to develop a Convolutional neural network (CNN) based Chl-a retrieval model utilizing both SAR data and optical data in Korean lakes. The model dataset was established by acquiring Chl-a concentration data and Sentinel-1/2 imagery from the Copernicus Open Access Hub. The CNN model trained on both optical and SAR data exhibited superior performance (R2 = 0.7992, RMSE = 10.3282 mg/m3, RPD = 2.2315) compared with the model trained exclusively on optical data. Moreover, SAR data exhibited moderate variable importance among all variables, demonstrating their efficacy as input variables for Chl-a concentration estimation. Furthermore, the CNN model estimated Chl-a concentrations with a spatial distribution that matched the observed spatial heterogeneity of Chl-a concentrations. These results are expected to serve as a foundation for future research on remote monitoring of Chl-a using such data.

28 June 2025 in Water
Verified
Journal Article
Open access
PDF file
SLSunmin Lee
Sunmin Lee
BJBongseok Jeong
Bongseok Jeong
DYDonghyeon Yoon
Donghyeon Yoon
JLJinhee Lee
Jinhee Lee
JLJeongho Lee
Jeongho Lee
Moung-Jin LeeMoung-Jin Lee
Moung-Jin Lee
Rapid rises in water levels due to heavy rainfall can lead to the accumulation of floating debris, posing significant challenges for both water quality and resource management. However, real-time monitoring of floating debris remains difficult due to the discrepancy between meteorological conditions and the timing of debris accumulation. To address this limitation, this study proposes an amplitude change detection (ACD) model based on time-series synthetic aperture radar (SAR) imagery, which is less affected by weather conditions. The model statistically distinguishes floating debris from open water based on their differing scattering characteristics. The ACD approach was applied to 18 pairs of Sentinel-1 SAR images acquired over Daecheong Reservoir from June to September 2024. A stringent type I error threshold (α < 1 × 10−8) was employed to ensure reliable detection. The results revealed a distinct cumulative effect, whereby the detected debris area increased immediately following rainfall events. A positive correlation was observed between 10-day cumulative precipitation and the debris-covered area. For instance, on 12 July, a floating debris area of 0.3828 km2 was detected, which subsequently expanded to 0.4504 km2 by 24 July. In contrast, on 22 August, when rainfall was negligible, no debris was detected (0 km2), indicating that precipitation was a key factor influencing the detection sensitivity. Comparative analysis with optical imagery further confirmed that floating debris tended to accumulate near artificial barriers and narrow channel regions. Overall, this study demonstrates that this spatial pattern suggests the potential to use detection results to estimate debris transport pathways and inform retrieval strategies.

15 May 2025 in Water Science & Technology
Verified
Journal Article
Open access
BJBongseok Jeong
Bongseok Jeong
HSHyunjoo Shin
Hyunjoo Shin
JSJihoon Shin
Jihoon Shin
YCYoonKyung Cha
YoonKyung Cha
Since rivers are adversely affected by harmful algal blooms (HABs), it is necessary to develop countermeasures through analyzing the relationship between environmental variables and HABs. This study focused on analyzing the connections between HABs and environmental variables in the middle reaches of the Nakdong River in South Korea. Partial least squares structural equation modeling (PLS-SEM) was used to identify those relationships. The study developed three different PLS-SEM models to investigate various aspects, including lagged effects of environmental factors, influences from upstream and tributaries, and interactions between genera of HABs. The results of the study revealed that the magnitude of HABs had the strongest relationships with nutrient concentrations, particularly 1 week prior to HABs measurement. Additionally, the magnitude of HABs showed stronger relationships with upstream nutrient concentrations compared with tributaries' nutrient concentrations. Furthermore, the dominant genus of HABs in the study area, Microcystis, showed significant relationships with temperature and nutrient concentrations. However, the study did not find significant relationships between Microcystis and other harmful cyanobacteria genera. The methodological framework provides valuable insight into the management of HABs. It allows for the analysis of multiple aspects of the relationships between environmental factors and HABs, which is crucial for effective water resource management.

31 December 2024 in GEO DATA
Verified
Journal Article
Open access
PDF file
BJBongseok Jeong
Bongseok Jeong
SLSunmin Lee
Sunmin Lee
Moung-Jin LeeMoung-Jin Lee
Moung-Jin Lee
Land cover provides crucial information related to biological geography, ecological climatology, and human activities. In the past, land cover mapping was performed based on visual interpretation, but it had limitations in terms of time and cost. Recently, it has become possible to create land cover maps with higher temporal resolution over wider areas using artificial intelligence-based models. The accuracy and reliability of AI model-based land cover maps increase with the amount of training data, but it is difficult to acquire large amounts of data due to the time required for label data annotation. In South Korea, the Environmental Geographic Information Service provides self-learning data consisting of aerial orthoimages and subdivision land cover classification level label data, making it possible to collect high-quality data. Therefore, this study examined the feasibility of self-learning data by building and evaluating a U-Net-based land cover classification model for waterfront areas using self-learning data. The trained model showed relatively low performance with an F-1 score of 0.61 for training data and 0.31 for test data. The model’s low performance is thought to be due to insufficient training caused by the large number of classification categories (34) and data imbalance between categories. Although the model performance using self-learning data was low, it is believed that model performance can be improved by grouping classification categories according to research purposes or resolving data imbalance through data augmentation techniques. Therefore, self-learning data is expected to be utilized in various studies using land cover.

30 June 2025 in GEO DATA
Verified
Journal Article
Open access
PDF file
BJBongseok Jeong
Bongseok Jeong
SLSunmin Lee
Sunmin Lee
Moung-Jin LeeMoung-Jin Lee
Moung-Jin Lee
Global attention to blue carbon, which absorbs significant amounts of carbon, has increased as an alternative solution for reducing greenhouse gases, a major concern in the international community. However, as blue carbon is threatened by land cover changes and sea-level rise, preliminary studies on assessment and restoration have been conducted for efficient management. While domestic research has focused on field surveys of carbon absorption capacity and storage, studies establishing spatial information systems have been limited. Therefore, this study developed blue carbon spatial information for the coastal watersheds of the four major rivers. Total carbon storage within watersheds was calculated based on data from field surveys of 21 domestic tidal flats and various tidal flat spatial information datasets (global distribution of tidal flat ecosystems, Subdivision land cover map, 1/25,000 coastal data and 2018 tidal flat data). Watershed health indices were calculated based on total carbon storage. The carbon storage per unit area was highest in the Han River coastal watershed, followed by Nakdong, Geum, and Yeongsan River Coastal Watersheds. Among the various tidal flat spatial information datasets applied, analysis revealed that the highest carbon storage values were found in Han (7,321,321.72 Mg C, Nakdong (197,943.19 Mg C), Geum (469,228.90 Mg C), and Yeongsan (2,849,389.2 Mg C) River Coastal Watersheds. The health index analysis comparing various tidal flat spatial information datasets indicated peak values in Han (54.12), Nakdong (82.92), Geum (75.40), and Yeongsan (59.58) River Coastal Watersheds. The blue carbon spatial information developed in this study can potentially support policy decision-making and future research utilizing spatial information, such as analyzing factors affecting blue carbon absorption rates.