Soojeong Jang

Soojeong Jang

Auto-generated profile, based on publicly available information

Soojeong Jang

Kyung Hee University Yongin South Korea

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
9 January 2024 in Space Weather
Verified
Journal Article
Open access
SJSe‐Heon Jeong
Se‐Heon Jeong
WLWoo Kyoung Lee
Woo Kyoung Lee
HKHyosub Kil
Hyosub Kil
SJSoojeong Jang
Soojeong Jang
JKJeong‐Heon Kim
Jeong‐Heon Kim
YKYoungsil Kwak
Youngsil Kwak
This study evaluates the performance of deep learning approach in the prediction of the ionospheric total electron content (TEC) during magnetically quiet periods. Two deep learning techniques, long short‐term memory (LSTM) and convolutional LSTM (ConvLSTM), are employed to predict TEC values 24 hr ahead in the vicinity of the Korean Peninsula (26.5°–40°N, 121°–134.5°E). The LSTM method predicts TEC at a single point based on time series of data at that point, whereas the ConvLSTM method simultaneously predicts TEC values at multiple points using spatiotemporal distribution of TEC. Both the LSTM and ConvLSTM models are trained using the complete regional TEC maps reconstructed by applying the Deep Convolutional Generative Adversarial Network–Poisson Blending (DCGAN‐PB) method to observed TEC data. The training period spans from 2002 to 2018, and the model performance is evaluated using 2019 data. Our results show that the ConvLSTM method outperforms the LSTM method, generating more reliable TEC maps with smaller root mean square errors when compared to the ground truth (DCGAN‐PB TEC maps). This outcome indicates that deep learning models can improve the prediction accuracy of TEC at a specific point by taking into account spatial information of TEC. We conclude that ConvLSTM is a reliable and efficient approach for the prompt ionospheric prediction.

1 April 2020 in The Astrophysical Journal
Verified
Journal Article
Closed access
JLJae-Ok Lee
Jae-Ok Lee
KCKyung-Suk Cho
Kyung-Suk Cho
KLKyoung-Sun Lee
Kyoung-Sun Lee
ICIl-Hyun Cho
Il-Hyun Cho
JLJunggi Lee
Junggi Lee
SJSoojeong Jang
Soojeong Jang
Understanding the formation of post-CME blobs, we investigate 2 blobs in the outer corona observed by LASCO-C2 and 34 blobs in the inner corona by K-Cor on 2017 September 10 from 17:11 to 18:58 UT. By visual inspection of the structure of a post-CME current sheet (CS) and its associated blobs, we find that the CS is well identified in the K-Cor and its radial lengths are nine times longer than lateral widths, indicating the CS is unstable to the linear tearing mode. The inner corona blobs can be classified into two groups: 27 blobs generated in the middle of the CS (Group 1) and 7 blobs occurred above the tips of it (Group 2). Their lateral widths are and , which is smaller than, or similar to, those of the CS. They have elongated shapes: ratios of lateral to radial widths are and , respectively. In the first group, only three blobs propagate above the tip of the CS while the others are located in the CS. In the second group, only two blobs have associations with those of outer corona in their temporal and spatial relationship and their initial heights are 1.81 and 1.95 R ☉, measured from the center of the Sun. The others cannot be identified in the outer corona. Our results first demonstrate that LASCO-C2 blobs could be generated by the tearing mode instability near the tips of post-CME CSs, similar to the magnetic reconnection process in the tail CS of Earth's magnetosphere.

8 August 2022 in Space Weather
Verified
Journal Article
Open access
SJSe‐Heon Jeong
Se‐Heon Jeong
WLWoo Kyoung Lee
Woo Kyoung Lee
SJSoojeong Jang
Soojeong Jang
HKHyosub Kil
Hyosub Kil
JKJeong‐Heon Kim
Jeong‐Heon Kim
BCByung-Kyu Choi
Byung-Kyu Choi
This study reconstructs total electron content (TEC) maps in the vicinity of the Korean Peninsula by employing a deep convolutional generative adversarial network and Poisson blending (DCGAN‐PB). Our interest is to rebuild small‐scale ionosphere structures on the TEC map in a local region where pronounced ionospheric structures, such as the equatorial ionization anomaly, are absent. The reconstructed regional TEC maps have a domain of 120°–135.5°E longitude and 25.5°–41°N latitude with 0.5° resolution. To achieve this, we first train a DCGAN model by using the International Reference Ionosphere‐based TEC maps from 2002 to 2019 (except for 2010 and 2014) as a training data set. Next, the trained DCGAN model generates synthetic complete TEC maps from observation‐based incomplete TEC maps. Final TEC maps are produced by blending of synthetic TEC maps with observed TEC data by PB. The performance of the DCGAN‐PB model is evaluated by testing the regeneration of the masked TEC observations in 2010 (solar minimum) and 2014 (solar maximum). Our results show that a good correlation between the masked and model‐generated TEC values is maintained even with a large percentage (∼80%) of masking. The performance of the DCGAN‐PB model is not sensitive to local time, solar activity, and magnetic activity. Thus, the DCGAN‐PB model can reconstruct fine ionospheric structures in regions where observations are sparse and distinguishing ionospheric structures are absent. This model can contribute to near real‐time monitoring of the ionosphere by immediately providing complete TEC maps.

1 October 2021 in The Astrophysical Journal Letters
Verified
Journal Article
Open access
PDF file
SJSoojeong Jang
Soojeong Jang
RKRyun-Young Kwon
Ryun-Young Kwon
JLJon A. Linker
Jon A. Linker
PRPete Riley
Pete Riley
GSGyungin Shin
Gyungin Shin
YKYeon-Han Kim
Yeon-Han Kim
We present, for the first time, a deep learning model that returns the three-dimensional (3D) coronal electron density from coronagraphic images. The intensity of coronagraphic observations arises from the Thomson scattering of photospheric light by the coronal electrons. We use MHD numerical simulations to obtain realistic 3D electron density and construct error-free training sets consisting of input (observation) and target (electron density) images. In the training sets, the input images are directly synthesized from the target 3D electron density by applying the Thomson scattering theory. The input and target images are in the form of latitude–longitude maps given at a radius, often referred to as synoptic maps. Using synoptic maps reduces a tomographic method to an image translation problem. We use pix2pixHD, one of the well-established supervised image translation methods and develop models for six selected heights: 2.0, 2.2, 2.5, 4.0, 6.0, and 12.0 solar radii. All six models have similar performance and the mean absolute percent error of the generated density images is less than 7% with respect to the ground-truth simulated data sets.

1 March 2020 in Journal of Astronomy and Space Sciences
Verified
Journal Article
Open access
PDF file
SPSa-Rah Park
Sa-Rah Park
HJHo-Cheol Jeon
Ho-Cheol Jeon
RKRok-Soon Kim
Rok-Soon Kim
JKJong-Hyeon Kim
Jong-Hyeon Kim
SKSeung-Jin Kim
Seung-Jin Kim
SJSoojeong Jang
Soojeong Jang
We have developed an algorithm for tracking coronal mass ejection (CME) propagation that allows us to estimate CME speed and its arrival time at Earth. The algorithm may be used either to forecast the CME’s arrival on the day of the forecast or to update the CME tracking information for the next day’s forecast. In our case study, we successfully tracked CME propagation using the algorithm based on g-values of interplanetary scintillation (IPS) observation provided by the Institute for Space- Earth Environmental Research (ISEE). We were able to forecast the arrival time (Δt = 0.30 h) and speed (Δv = 20 km/s) of a CME event on October 2, 2000. From the CME-interplanetary CME (ICME) pairs provided by Cane & Richardson (2003), we selected 50 events to evaluate the algorithm’s forecast capability. Average errors for arrival time and speed were 11.14 h and 310 km/s, respectively. Results demonstrated that g-values obtained continuously from any single station observation were able to be used as a proxy for CME speed. Therefore, our algorithm may give stable daily forecasts of CME position and speed during propagation in the region of 0.2–1 AU using the IPS g-values, even if IPS velocity observations are insufficient. We expect that this algorithm may be widely accepted for use in space weather forecasting in the near future.