R. Mukesh, Sarat C. Dass, M. Vijay, S. Kiruthiga, Vijanth Sagayam
{"title":"LSTM 和 OKSM 对 2023 年发生 M 级太阳耀斑期间电离层 TEC 的预测","authors":"R. Mukesh, Sarat C. Dass, M. Vijay, S. Kiruthiga, Vijanth Sagayam","doi":"10.1007/s10509-024-04290-x","DOIUrl":null,"url":null,"abstract":"<div><p>Advancements in space weather forecasting have become crucial for understanding and mitigating the impacts of solar activity on Earth’s ionosphere. This research focuses on the prediction of Total Electron Content (TEC) during M-class solar flare events in 2023. TEC is a vital parameter for satellite communications and navigation, making accurate forecasting imperative. Two prediction models, Long Short-Term Memory (LSTM) neural networks and Surrogate Models based on Ordinary Kriging (OKSM), are employed. LSTM, known for capturing temporal dependencies, is contrasted with OKSM, a geostatistical interpolation technique capturing spatial autocorrelation. The study utilizes TEC measurements from the Hyderabad (HYDE) GPS station for model training and evaluation along with solar and geomagnetic parameters. The performance metrics for both models across various solar flare dates are measured using Root Mean Square Error (RMSE), Normalized RMSE, Correlation Coefficient (CC), and Symmetric Mean Absolute Percentage Error(sMAPE). The research interprets the results, highlighting the strengths and limitations of each model. Notable findings include LSTM’s proficiency in capturing temporal variations and OKSM’s unique spatial perspective. Different solar flare intensities are analyzed separately, demonstrating the model’s adaptability to varying space weather conditions. The average performance metrics during M 4.65 SF events for the OKSM model, in terms of Root Mean Square Error is 5.61, Normalized RMSE is 0.14, Correlation Coefficient is 0.9813, and Symmetric Mean Absolute Percentage Error is 14.90. Similarly, for LSTM, the corresponding averages are 10.03, 0.24, 0.9313, and 28.64. The research contributes valuable insights into the predictive capabilities of LSTM and OKSM for TEC during solar flare events. The outcomes aid in understanding the applicability of machine learning and geostatistical techniques in space weather prediction. As society’s reliance on technology susceptible to space weather effects grows, this research is pivotal for enhancing space weather forecasts and ensuring the robustness of critical technological infrastructure on Earth.</p></div>","PeriodicalId":8644,"journal":{"name":"Astrophysics and Space Science","volume":null,"pages":null},"PeriodicalIF":1.8000,"publicationDate":"2024-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Prediction of ionospheric TEC by LSTM and OKSM during M class solar flares occurred during the year 2023\",\"authors\":\"R. Mukesh, Sarat C. Dass, M. Vijay, S. Kiruthiga, Vijanth Sagayam\",\"doi\":\"10.1007/s10509-024-04290-x\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Advancements in space weather forecasting have become crucial for understanding and mitigating the impacts of solar activity on Earth’s ionosphere. This research focuses on the prediction of Total Electron Content (TEC) during M-class solar flare events in 2023. TEC is a vital parameter for satellite communications and navigation, making accurate forecasting imperative. Two prediction models, Long Short-Term Memory (LSTM) neural networks and Surrogate Models based on Ordinary Kriging (OKSM), are employed. LSTM, known for capturing temporal dependencies, is contrasted with OKSM, a geostatistical interpolation technique capturing spatial autocorrelation. The study utilizes TEC measurements from the Hyderabad (HYDE) GPS station for model training and evaluation along with solar and geomagnetic parameters. The performance metrics for both models across various solar flare dates are measured using Root Mean Square Error (RMSE), Normalized RMSE, Correlation Coefficient (CC), and Symmetric Mean Absolute Percentage Error(sMAPE). The research interprets the results, highlighting the strengths and limitations of each model. Notable findings include LSTM’s proficiency in capturing temporal variations and OKSM’s unique spatial perspective. Different solar flare intensities are analyzed separately, demonstrating the model’s adaptability to varying space weather conditions. The average performance metrics during M 4.65 SF events for the OKSM model, in terms of Root Mean Square Error is 5.61, Normalized RMSE is 0.14, Correlation Coefficient is 0.9813, and Symmetric Mean Absolute Percentage Error is 14.90. Similarly, for LSTM, the corresponding averages are 10.03, 0.24, 0.9313, and 28.64. The research contributes valuable insights into the predictive capabilities of LSTM and OKSM for TEC during solar flare events. The outcomes aid in understanding the applicability of machine learning and geostatistical techniques in space weather prediction. As society’s reliance on technology susceptible to space weather effects grows, this research is pivotal for enhancing space weather forecasts and ensuring the robustness of critical technological infrastructure on Earth.</p></div>\",\"PeriodicalId\":8644,\"journal\":{\"name\":\"Astrophysics and Space Science\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.8000,\"publicationDate\":\"2024-03-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Astrophysics and Space Science\",\"FirstCategoryId\":\"101\",\"ListUrlMain\":\"https://link.springer.com/article/10.1007/s10509-024-04290-x\",\"RegionNum\":4,\"RegionCategory\":\"物理与天体物理\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ASTRONOMY & ASTROPHYSICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Astrophysics and Space Science","FirstCategoryId":"101","ListUrlMain":"https://link.springer.com/article/10.1007/s10509-024-04290-x","RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ASTRONOMY & ASTROPHYSICS","Score":null,"Total":0}
Prediction of ionospheric TEC by LSTM and OKSM during M class solar flares occurred during the year 2023
Advancements in space weather forecasting have become crucial for understanding and mitigating the impacts of solar activity on Earth’s ionosphere. This research focuses on the prediction of Total Electron Content (TEC) during M-class solar flare events in 2023. TEC is a vital parameter for satellite communications and navigation, making accurate forecasting imperative. Two prediction models, Long Short-Term Memory (LSTM) neural networks and Surrogate Models based on Ordinary Kriging (OKSM), are employed. LSTM, known for capturing temporal dependencies, is contrasted with OKSM, a geostatistical interpolation technique capturing spatial autocorrelation. The study utilizes TEC measurements from the Hyderabad (HYDE) GPS station for model training and evaluation along with solar and geomagnetic parameters. The performance metrics for both models across various solar flare dates are measured using Root Mean Square Error (RMSE), Normalized RMSE, Correlation Coefficient (CC), and Symmetric Mean Absolute Percentage Error(sMAPE). The research interprets the results, highlighting the strengths and limitations of each model. Notable findings include LSTM’s proficiency in capturing temporal variations and OKSM’s unique spatial perspective. Different solar flare intensities are analyzed separately, demonstrating the model’s adaptability to varying space weather conditions. The average performance metrics during M 4.65 SF events for the OKSM model, in terms of Root Mean Square Error is 5.61, Normalized RMSE is 0.14, Correlation Coefficient is 0.9813, and Symmetric Mean Absolute Percentage Error is 14.90. Similarly, for LSTM, the corresponding averages are 10.03, 0.24, 0.9313, and 28.64. The research contributes valuable insights into the predictive capabilities of LSTM and OKSM for TEC during solar flare events. The outcomes aid in understanding the applicability of machine learning and geostatistical techniques in space weather prediction. As society’s reliance on technology susceptible to space weather effects grows, this research is pivotal for enhancing space weather forecasts and ensuring the robustness of critical technological infrastructure on Earth.
期刊介绍:
Astrophysics and Space Science publishes original contributions and invited reviews covering the entire range of astronomy, astrophysics, astrophysical cosmology, planetary and space science and the astrophysical aspects of astrobiology. This includes both observational and theoretical research, the techniques of astronomical instrumentation and data analysis and astronomical space instrumentation. We particularly welcome papers in the general fields of high-energy astrophysics, astrophysical and astrochemical studies of the interstellar medium including star formation, planetary astrophysics, the formation and evolution of galaxies and the evolution of large scale structure in the Universe. Papers in mathematical physics or in general relativity which do not establish clear astrophysical applications will no longer be considered.
The journal also publishes topically selected special issues in research fields of particular scientific interest. These consist of both invited reviews and original research papers. Conference proceedings will not be considered. All papers published in the journal are subject to thorough and strict peer-reviewing.
Astrophysics and Space Science features short publication times after acceptance and colour printing free of charge.