11:15 〜 11:30
[PEM09-09] Deep Learning Modeling of the Plasma Bubbles in Equatorial to Midlatitude Ionosphere From GNSS-Derived ROTI Maps
キーワード：Plasma Bubbles, Deep learning, ROTI map
The equatorial ionospheric irregularity, commonly known as the equatorial plasma bubble (EPB), is an ionospheric phenomenon that frequently occurs at the magnetic equator and at low latitudes. This phenomenon is characterized by a relatively lower electron density in parts of the ionosphere than in the surrounding regions. It causes rapid fluctuations in the amplitude and phase of radio signals, which are crucial for satellite communications and navigation systems. Therefore, it is a hot topic in ionosphere research.
Over the past 20 years, based on the global navigation satellite system (GNSS) amplitude scintillation S4 and total electron content index of irregular occurrence rate (ROTI) instructions and other measurement data has been widely used to study the morphological characteristics of ionospheric anomalies, the incidence of statistical anomalies in different areas, and forecasting the occurrence of ionospheric plasma irregular and flicker, etc.
The revolution of computer hardware, such as massive graphic processing units (GPUs) and the explosion of the amount of available data resurrect the field of artificial intelligence (AI) in recent years. Many AI algorithms, particularly deep learning that automatically combines the feature extraction and learning together through deep network structures, have been developed and achieved unprecedented performance in many fields, such as internet services, self-driving, health care, and gaming. In the field of space physics, especially in the upper atmosphere and the ionosphere, abundant global observation data create conditions for "deep learning" references in the upper atmosphere and the ionosphere, providing it with sufficient data to learn the characteristics of the upper atmosphere and the ionosphere.
Based on this, In this study, we try to use the global ROTI map data in the past 30 years, and use the deep learning algorithm to conduct AI modeling of the ROTI map, so as to systematically study and analyze the ionospheric irregularities, and study some basic characteristics of plasma bubbles, such as spatial and temporal distribution, morphological scale and solar activity influence. In addition, the differences in the incidence of ionospheric irregularities in different solar activities, geomagnetic activities and different seasons are studied and analyzed, so as to help us better understand the characteristics of ionospheric irregularities, and then better understand the generation mechanism of ionospheric irregularities.
Over the past 20 years, based on the global navigation satellite system (GNSS) amplitude scintillation S4 and total electron content index of irregular occurrence rate (ROTI) instructions and other measurement data has been widely used to study the morphological characteristics of ionospheric anomalies, the incidence of statistical anomalies in different areas, and forecasting the occurrence of ionospheric plasma irregular and flicker, etc.
The revolution of computer hardware, such as massive graphic processing units (GPUs) and the explosion of the amount of available data resurrect the field of artificial intelligence (AI) in recent years. Many AI algorithms, particularly deep learning that automatically combines the feature extraction and learning together through deep network structures, have been developed and achieved unprecedented performance in many fields, such as internet services, self-driving, health care, and gaming. In the field of space physics, especially in the upper atmosphere and the ionosphere, abundant global observation data create conditions for "deep learning" references in the upper atmosphere and the ionosphere, providing it with sufficient data to learn the characteristics of the upper atmosphere and the ionosphere.
Based on this, In this study, we try to use the global ROTI map data in the past 30 years, and use the deep learning algorithm to conduct AI modeling of the ROTI map, so as to systematically study and analyze the ionospheric irregularities, and study some basic characteristics of plasma bubbles, such as spatial and temporal distribution, morphological scale and solar activity influence. In addition, the differences in the incidence of ionospheric irregularities in different solar activities, geomagnetic activities and different seasons are studied and analyzed, so as to help us better understand the characteristics of ionospheric irregularities, and then better understand the generation mechanism of ionospheric irregularities.