Youngseok Song
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Youngseok Song has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
Soon Jee Seol
Department of Earth Resources and Environmental Engineering, Hanyang University , Seongdong-gu, Seoul 04763, Republic of Korea
3 shared publicationsSungmyung Bae
3 shared publications- Yonggyu Choi
Department of Earth Resources and Environmental Engineering (EREE), Hanyang University, Seoul, South Korea
3 shared publications - Cheonhwan Lee2 shared publications
Joongmoo Byun
2 shared publications- Joongmoo Byun
Hanyang University
2 shared publications
Latest Publications
- 31 August 2025
- Published by The Korean Society of Climate Change Research in Journal of Climate Change Research
- Vol. 16 (4) , 667-684
Structural shifts in climate technology research: An integrated Revealed Comparative Advantage - Latent Dirichlet Allocation analysis of IPCC influence and national comparative advantages
This study identifies global research trends in climate technology and analyzes structural characteristics of national research focus using an integrated framework combining topic modeling (LDA) and relative comparative advantage (RCA) indices. Climate change response requires coordinated international research efforts, yet systematic analysis of national portfolios alignment with global policy agendas is limited. By examining 440,000 papers published between 1990 ~ 2024 from Web of Science, we measure topic-specific specializations across six major countries-South Korea, US, China, Germany, UK, and Australia - comparing structural patterns of Intergovernmental Panel on Climate Change (IPCC) - cited and non-cited literature. Our methodology employs RCA analysis for national comparative advantages, LDA topic modeling for thematic structures, and UMAP-based dimensionality reduction to detect structural transition points. The analysis reveals strong South Korean specialization in techno-centric fields like materials science, electrochemistry, and thermodynamics (RCA > 2.0) but weak representation in social sciences and governance (RCA < 0.3). Conversely, the US and UK demonstrate strong engagement in policy, public health, and social justice research, reflecting balanced portfolios. Using UMAP clustering, we identify distinct thematic trajectories and transition points (1991 ~ 1994, 2000 ~ 2001, 2005 ~ 2006), illustrating the increasing diversification of climate research, with IPCC reports driving topic shifts. These findings underscore Korea’s need to strategically realign its portfolio by integrating social-technological approaches and enhancing engagement in internationally coordinated research domains.
- 31 August 2025
- Published by The Korean Society of Climate Change Research in Journal of Climate Change Research
- Vol. 16 (4) , 667-684
Structural shifts in climate technology research: An integrated Revealed Comparative Advantage - Latent Dirichlet Allocation analysis of IPCC influence and national comparative advantages
This study identifies global research trends in climate technology and analyzes structural characteristics of national research focus using an integrated framework combining topic modeling (LDA) and relative comparative advantage (RCA) indices. Climate change response requires coordinated international research efforts, yet systematic analysis of national portfolios alignment with global policy agendas is limited. By examining 440,000 papers published between 1990 ~ 2024 from Web of Science, we measure topic-specific specializations across six major countries-South Korea, US, China, Germany, UK, and Australia - comparing structural patterns of Intergovernmental Panel on Climate Change (IPCC) - cited and non-cited literature. Our methodology employs RCA analysis for national comparative advantages, LDA topic modeling for thematic structures, and UMAP-based dimensionality reduction to detect structural transition points. The analysis reveals strong South Korean specialization in techno-centric fields like materials science, electrochemistry, and thermodynamics (RCA > 2.0) but weak representation in social sciences and governance (RCA < 0.3). Conversely, the US and UK demonstrate strong engagement in policy, public health, and social justice research, reflecting balanced portfolios. Using UMAP clustering, we identify distinct thematic trajectories and transition points (1991 ~ 1994, 2000 ~ 2001, 2005 ~ 2006), illustrating the increasing diversification of climate research, with IPCC reports driving topic shifts. These findings underscore Korea’s need to strategically realign its portfolio by integrating social-technological approaches and enhancing engagement in internationally coordinated research domains.
- 30 June 2024
- Published by The Korean Society of Climate Change Research in Journal of Climate Change Research
- Vol. 15 (3) , 289-305
Analyzing trends in climate technology cooperation in international organizations and the Korean government using natural language processing
This study investigates temporal trends and changes in climate technology cooperation through natural language processing of international organizational news and Korean policy briefings. Documents from 2019 to 2023 were collected based on keywords related to climate change from UNFCCC, CTCN, GCF, and Korean policy briefings. Web scraping, text mining, and community detection-based clustering were performed using packages in R and Python, and quadratic assignment procedure correlation analysis used Ucinet programs. First, this study identified trends and patterns of climate technology cooperation by period, as shown in the news from international organizations (UNFCCC, CTCN, GCF) and Korean policy briefings. UNFCCC news, influenced by COVID-19, shows significant topic distribution changes, reflecting global issue sensitivity. The GCF news maintains stability in climate technology and funding trends. Korean policy briefings demonstrate consistent topic distribution, emphasizing Korean climate policy discussions. Text mining reveals nuanced organizational differences in climate change goals. UNFCCC and CTCN prioritize SDG-focused initiatives, while GCF focuses on financial support for developing countries. Second, this study comprehensively compared and analyzed the climate technology cooperation trends in international organizational news and Korean policy briefings and classified the topics into reduction, adaptation, and general to identify the importance of climate technology cooperation trends. Expert discussions highlight mitigation coverage in Korean briefings and GCF. The GCF news specializes in limited topics with a consistent reporting pattern, while UNFCCC leads diverse climate initiatives. This study unveils news coverage similarities and differences between the Korean government and international organizations, offering valuable insights for global cooperation and policy formulation.
- 30 June 2024
- Published by The Korean Society of Climate Change Research in Journal of Climate Change Research
- Vol. 15 (3) , 289-305
Analyzing trends in climate technology cooperation in international organizations and the Korean government using natural language processing
This study investigates temporal trends and changes in climate technology cooperation through natural language processing of international organizational news and Korean policy briefings. Documents from 2019 to 2023 were collected based on keywords related to climate change from UNFCCC, CTCN, GCF, and Korean policy briefings. Web scraping, text mining, and community detection-based clustering were performed using packages in R and Python, and quadratic assignment procedure correlation analysis used Ucinet programs. First, this study identified trends and patterns of climate technology cooperation by period, as shown in the news from international organizations (UNFCCC, CTCN, GCF) and Korean policy briefings. UNFCCC news, influenced by COVID-19, shows significant topic distribution changes, reflecting global issue sensitivity. The GCF news maintains stability in climate technology and funding trends. Korean policy briefings demonstrate consistent topic distribution, emphasizing Korean climate policy discussions. Text mining reveals nuanced organizational differences in climate change goals. UNFCCC and CTCN prioritize SDG-focused initiatives, while GCF focuses on financial support for developing countries. Second, this study comprehensively compared and analyzed the climate technology cooperation trends in international organizational news and Korean policy briefings and classified the topics into reduction, adaptation, and general to identify the importance of climate technology cooperation trends. Expert discussions highlight mitigation coverage in Korean briefings and GCF. The GCF news specializes in limited topics with a consistent reporting pattern, while UNFCCC leads diverse climate initiatives. This study unveils news coverage similarities and differences between the Korean government and international organizations, offering valuable insights for global cooperation and policy formulation.
- 31 May 2022
- Published by Oxford University Press (OUP) in Geophysical Journal International
- Vol. 231 (1) , 256-268
Imaging subsurface structure over the Xiadian fault using P waves extracted from urban traffic noise
SUMMARY: The imaging of subsurface structures is an essential task in subsurface engineering projects; it provides information regarding the locations of active faults and layer boundaries. Among the methods available for imaging of subsurface structures, the body wave imaging method using urban traffic noise has recently attracted attention because it permits continuous measurement at low cost in urban areas. However, because the urban traffic noise signal used for imaging on the engineering scale has characteristics that differ from the ambient noise used on the crustal scale, the conventional crustal-scale data processing workflow should be modified through systematic data analysis. In this study, traffic noise sources were systematically analysed using field data obtained over the Xiadian fault in Hebei province, China. The traffic noise signals were recorded in various patterns because of diverse incoming directions and show marked amplitude changes depending on time of recording. The overlapping signals originating from opposite directions generate spurious events and noise in the seismic interferometry images; constant processing parameters cannot respond to the large amplitude changes. In this study, to remove surface waves with markedly changing amplitude, we applied actively varying threshold values to each set of traces using the moving average of amplitude changes within the trace. In addition, the signals originating from diverse directions were separated into negative and positive slopes through the f–k filter; the interference generated by overlapping signals was minimized by applying data processing (e.g. median filtering and high amplitude removal) separately to the negative and positive slopes of each simultaneously acquired trace gather. Due to the modified data processing workflow, most spurious events were successfully suppressed in the final stacked image compared with those produced using the conventional data processing workflow, and reflections were imaged more clearly. Fault spatial locations and layer boundary depth variation in the final image obtained by the modified processing workflow were similar to those reported in previous studies.
- 31 May 2022
- Published by Oxford University Press (OUP) in Geophysical Journal International
- Vol. 231 (1) , 256-268
Imaging subsurface structure over the Xiadian fault using P waves extracted from urban traffic noise
SUMMARY: The imaging of subsurface structures is an essential task in subsurface engineering projects; it provides information regarding the locations of active faults and layer boundaries. Among the methods available for imaging of subsurface structures, the body wave imaging method using urban traffic noise has recently attracted attention because it permits continuous measurement at low cost in urban areas. However, because the urban traffic noise signal used for imaging on the engineering scale has characteristics that differ from the ambient noise used on the crustal scale, the conventional crustal-scale data processing workflow should be modified through systematic data analysis. In this study, traffic noise sources were systematically analysed using field data obtained over the Xiadian fault in Hebei province, China. The traffic noise signals were recorded in various patterns because of diverse incoming directions and show marked amplitude changes depending on time of recording. The overlapping signals originating from opposite directions generate spurious events and noise in the seismic interferometry images; constant processing parameters cannot respond to the large amplitude changes. In this study, to remove surface waves with markedly changing amplitude, we applied actively varying threshold values to each set of traces using the moving average of amplitude changes within the trace. In addition, the signals originating from diverse directions were separated into negative and positive slopes through the f–k filter; the interference generated by overlapping signals was minimized by applying data processing (e.g. median filtering and high amplitude removal) separately to the negative and positive slopes of each simultaneously acquired trace gather. Due to the modified data processing workflow, most spurious events were successfully suppressed in the final stacked image compared with those produced using the conventional data processing workflow, and reflections were imaged more clearly. Fault spatial locations and layer boundary depth variation in the final image obtained by the modified processing workflow were similar to those reported in previous studies.
- Open Access
- 28 March 2022
- Published by Copernicus GmbH
P- and S-wave arrival picking and epicentral distance estimation of earthquakes using convolutional neural networks
In recent years, machine learning techniques have been widely applied in seismological data processing such as seismic event detection, phase picking, location, magnitude estimation, and further data analysis for determining source mechanisms. Especially in earthquake location, deep learning methods are used to reduce location errors compared to conventional algorithms.In this study, we present a deep learning based epicentral distance estimation with two separate models using seismic data from two stations as input data. The first model is the P- and S-wave arrival time picking model and the second is the epicentral distance estimation model. Since the traditional epicentral distance estimation methods uses the difference in arrival times between P- and S-waves, the P- and S-wave arrival times were first predicted from three-component seismic data so that this information could be directly used as the next input data. This picking information is used as input data along with the station location in the epicentral distance estimation model to output the final epicentral distance. Since this method uses data from two stations, it has higher accuracy than epicentral distance estimation using data from a single station.The P- and S-wave arrival time picking model was modified by referring to the ResUNet (Diakogiannis et al., 2020) structure to improve performance based on the seismic detection and phase picking model from the three-component acceleration data developed by Mousavi et al. (2020). This modified model performs feature extraction for P- and S-phase picking and includes a residual block and skip connection. The model for estimating the distance from the epicenter was constructed using a basic artificial neural network (ANN) architecture. As input data, a total of eight features were used by adding six combinations of the difference in arrival times of P-wave and S-wave in each component of the two stations and two values of the latitude and longitude difference between two stations. The ANN architecture consists of four hidden layers and the epicentral distances of the two stations are final output.The STEAD data were used as training data and test data. The STEAD is a seismogram dataset recorded from about 450,000 global earthquakes, and among them, data with magnitudes greater than 2.5 and epicentral distances less than 400 km were selected and used. As a result of applying the trained model to the test data, the mean absolute error of the predicted epicentral distance was 6.5 km, which showed improved performance compared to the previous results. Also, since this method uses six time-differences as input data, it can provide more robust results even in the presence of random noise at the picked times.
- 28 March 2022
- Published by Copernicus GmbH
Open AccessP- and S-wave arrival picking and epicentral distance estimation of earthquakes using convolutional neural networks
In recent years, machine learning techniques have been widely applied in seismological data processing such as seismic event detection, phase picking, location, magnitude estimation, and further data analysis for determining source mechanisms. Especially in earthquake location, deep learning methods are used to reduce location errors compared to conventional algorithms.In this study, we present a deep learning based epicentral distance estimation with two separate models using seismic data from two stations as input data. The first model is the P- and S-wave arrival time picking model and the second is the epicentral distance estimation model. Since the traditional epicentral distance estimation methods uses the difference in arrival times between P- and S-waves, the P- and S-wave arrival times were first predicted from three-component seismic data so that this information could be directly used as the next input data. This picking information is used as input data along with the station location in the epicentral distance estimation model to output the final epicentral distance. Since this method uses data from two stations, it has higher accuracy than epicentral distance estimation using data from a single station.The P- and S-wave arrival time picking model was modified by referring to the ResUNet (Diakogiannis et al., 2020) structure to improve performance based on the seismic detection and phase picking model from the three-component acceleration data developed by Mousavi et al. (2020). This modified model performs feature extraction for P- and S-phase picking and includes a residual block and skip connection. The model for estimating the distance from the epicenter was constructed using a basic artificial neural network (ANN) architecture. As input data, a total of eight features were used by adding six combinations of the difference in arrival times of P-wave and S-wave in each component of the two stations and two values of the latitude and longitude difference between two stations. The ANN architecture consists of four hidden layers and the epicentral distances of the two stations are final output.The STEAD data were used as training data and test data. The STEAD is a seismogram dataset recorded from about 450,000 global earthquakes, and among them, data with magnitudes greater than 2.5 and epicentral distances less than 400 km were selected and used. As a result of applying the trained model to the test data, the mean absolute error of the predicted epicentral distance was 6.5 km, which showed improved performance compared to the previous results. Also, since this method uses six time-differences as input data, it can provide more robust results even in the presence of random noise at the picked times.
- Open Access
- 28 March 2022
- Published by Copernicus GmbH
Machine learning-based attenuation of steeply dipping events of seismic reflection image beneath the Korean Peninsula
Seismic reflection images derived by ambient-noise seismic interferometry (SI) can show subsurface structures without active sources. To image and interpret the upper mantle structures and tectonic boundaries beneath the southern part of Korean Peninsula, we applied SI method to seismic ambient noise data recorded at 119 seismic stations on the Korean Peninsula in 2014 (from the seismic network of the Korean Meteorological Administration). The factor that makes interpretation difficult is the steeply dipping events in reflection images. Most of these events of apparent steeply dips show as true reflection events from steep geologic boundaries. Therefore, we need to attenuate these events to interpret true reflection events. These events overlap many times. Also, the value of the slope has several values close to half of the Rayleigh waves or P waves. To attenuate these events with these complex features, we used machine learning techniques. We attenuated our steeply dipping events by applying the Extraction of diffractions method. As the steeply dipping events are attenuated, horizontal events were strengthened, and noises were attenuated. We can more clearly identify the reflection events of the Moho discontinuity and the lithosphere/asthenosphere (LAB) boundary near the two-way reflection times of 7-11 s and 17-22 s respectively.
- 28 March 2022
- Published by Copernicus GmbH
Open AccessMachine learning-based attenuation of steeply dipping events of seismic reflection image beneath the Korean Peninsula
Seismic reflection images derived by ambient-noise seismic interferometry (SI) can show subsurface structures without active sources. To image and interpret the upper mantle structures and tectonic boundaries beneath the southern part of Korean Peninsula, we applied SI method to seismic ambient noise data recorded at 119 seismic stations on the Korean Peninsula in 2014 (from the seismic network of the Korean Meteorological Administration). The factor that makes interpretation difficult is the steeply dipping events in reflection images. Most of these events of apparent steeply dips show as true reflection events from steep geologic boundaries. Therefore, we need to attenuate these events to interpret true reflection events. These events overlap many times. Also, the value of the slope has several values close to half of the Rayleigh waves or P waves. To attenuate these events with these complex features, we used machine learning techniques. We attenuated our steeply dipping events by applying the Extraction of diffractions method. As the steeply dipping events are attenuated, horizontal events were strengthened, and noises were attenuated. We can more clearly identify the reflection events of the Moho discontinuity and the lithosphere/asthenosphere (LAB) boundary near the two-way reflection times of 7-11 s and 17-22 s respectively.
- Open Access
- 27 March 2022
- Published by Copernicus GmbH
Seismic Intensity Estimation using Machine Learning for on-site Earthquake Early Warning (EEW)
Earthquake Early Warning System (EEW) is a technology that calculates earthquake parameter using P-wave that arrives earlier and warns the expected damage area before the arrival of destructive S wave. Therefore, many countries are operating EEW to mitigate damage from earthquake shaking. Especially an on-site EEW is drawn attention as it can reduce blind zones due to using only a single or minimum station. In the on-site EEW, it is important to quickly predict the seismic intensity, which indicates the degree of ground damage, response of structures and ground shaking felt by people at a given location, rather than information on the magnitude or distance of the earthquake.In this study, we suggest a machine learning (ML) model that can directly estimate the seismic intensity scale from initial P-waveforms of three-component acceleration data measured at a single station. We used 1D-Convolutional Neural Networks (1D-CNN), which have been shown good performance in signal processing of speech and medical data which are similar to earthquake signals. K-Net and KiK-net datasets, recorded at stations in Japan, were used for training the ML model. Since the amount of data is enough and all of data are labeled with Japan Meteorological Agency Seismic Intensity Scale (IJMA), the datasets were used as training data in this study. The developed model produced fast and accurate results using only the three-component acceleration field data at a single station.In order to test applicability of the trained model to the new dataset acquired from other regions, the trained model was applied to the STEAD data which were recorded at stations distributed globally. When the trained model was applied to STEAD data directly, the prediction results were worse than those of K-Net and KiK-net data. The reason is that the characteristics of the ground and waveforms are different depending on the region. Therefore, to solve this problem, transfer learning was applied, and only the parameters of a fully connected layer of pretrained ML model were fine-tuned using small number of both labeled target dataset and training dataset used for pretraining. Moreover, by considering the imbalance problem of the training data for transfer learning, it was able to obtain better prediction results. Ultimately, this study shows the pretrained model with a specific region dataset can provide reasonable prediction of seismic intensity to new dataset acquired from other regions using transfer learning.
- 27 March 2022
- Published by Copernicus GmbH
Open AccessSeismic Intensity Estimation using Machine Learning for on-site Earthquake Early Warning (EEW)
Earthquake Early Warning System (EEW) is a technology that calculates earthquake parameter using P-wave that arrives earlier and warns the expected damage area before the arrival of destructive S wave. Therefore, many countries are operating EEW to mitigate damage from earthquake shaking. Especially an on-site EEW is drawn attention as it can reduce blind zones due to using only a single or minimum station. In the on-site EEW, it is important to quickly predict the seismic intensity, which indicates the degree of ground damage, response of structures and ground shaking felt by people at a given location, rather than information on the magnitude or distance of the earthquake.In this study, we suggest a machine learning (ML) model that can directly estimate the seismic intensity scale from initial P-waveforms of three-component acceleration data measured at a single station. We used 1D-Convolutional Neural Networks (1D-CNN), which have been shown good performance in signal processing of speech and medical data which are similar to earthquake signals. K-Net and KiK-net datasets, recorded at stations in Japan, were used for training the ML model. Since the amount of data is enough and all of data are labeled with Japan Meteorological Agency Seismic Intensity Scale (IJMA), the datasets were used as training data in this study. The developed model produced fast and accurate results using only the three-component acceleration field data at a single station.In order to test applicability of the trained model to the new dataset acquired from other regions, the trained model was applied to the STEAD data which were recorded at stations distributed globally. When the trained model was applied to STEAD data directly, the prediction results were worse than those of K-Net and KiK-net data. The reason is that the characteristics of the ground and waveforms are different depending on the region. Therefore, to solve this problem, transfer learning was applied, and only the parameters of a fully connected layer of pretrained ML model were fine-tuned using small number of both labeled target dataset and training dataset used for pretraining. Moreover, by considering the imbalance problem of the training data for transfer learning, it was able to obtain better prediction results. Ultimately, this study shows the pretrained model with a specific region dataset can provide reasonable prediction of seismic intensity to new dataset acquired from other regions using transfer learning.