Han Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Han Wu has a total of 42 co-authors. Here are the top co-authors by number of shared publications:
- Hanxian Fang
College of Advanced Interdisciplinary Studies National University of Defense Technology Changsha China
5 shared publications - Chao Xiao
College of Advanced Interdisciplinary Studies National University of Defense Technology Changsha China
3 shared publications - Chenghao Li
College of Meteorology and Oceanography National University of Defense Technology Changsha China
3 shared publications - Die Duan
College of Advanced Interdisciplinary Studies National University of Defense Technology Changsha China
3 shared publications - Ganming Ren
College of Advanced Interdisciplinary Studies National University of Defense Technology Changsha China
3 shared publications - Hongtao Huang
College of Advanced Interdisciplinary Studies National University of Defense Technology Changsha China
3 shared publications
Latest Publications
- 14 November 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 131 (11)
Extrapolating Meteor Radar Winds to 110 and 120 km: Implications for Mid‐Latitude Sporadic E Layers
Sporadic E layers (Es layers) are irregular plasma structures located at 90–120 km altitude within the ionosphere, exerting a significant impact on shortwave communications. Reliable wind measurements at 110 and 120 km are essential for understanding and monitoring Es layers; however, meteor radars often lack continuous observations at these altitudes. Here, we develop a satellite‐trained encoder–decoder fully connected wind model (ED‐FCWM) to extrapolate meteor‐radar winds from 90 to 98 km to higher altitudes by leveraging relationships learned from satellite wind profiles. The extrapolated winds substantially outperform HWM2014, with the average root‐mean‐square error (RMSE) significantly reduced by more than 19%. Furthermore, analysis of two major sudden stratospheric warming (SSW) events reveals that the diurnal tidal component of the extrapolated zonal wind displays approximately 10‐day oscillations, which align well with variations observed in the 24‐hr component of the concurrent Es layers. These results demonstrate that satellite‐trained wind extrapolation can effectively bridge the observational gap above meteor radar altitudes, contributing to a better understanding of the physical mechanisms of Es layers and enhancing forecasting capabilities.
- 14 November 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 131 (11)
Extrapolating Meteor Radar Winds to 110 and 120 km: Implications for Mid‐Latitude Sporadic E Layers
Sporadic E layers (Es layers) are irregular plasma structures located at 90–120 km altitude within the ionosphere, exerting a significant impact on shortwave communications. Reliable wind measurements at 110 and 120 km are essential for understanding and monitoring Es layers; however, meteor radars often lack continuous observations at these altitudes. Here, we develop a satellite‐trained encoder–decoder fully connected wind model (ED‐FCWM) to extrapolate meteor‐radar winds from 90 to 98 km to higher altitudes by leveraging relationships learned from satellite wind profiles. The extrapolated winds substantially outperform HWM2014, with the average root‐mean‐square error (RMSE) significantly reduced by more than 19%. Furthermore, analysis of two major sudden stratospheric warming (SSW) events reveals that the diurnal tidal component of the extrapolated zonal wind displays approximately 10‐day oscillations, which align well with variations observed in the 24‐hr component of the concurrent Es layers. These results demonstrate that satellite‐trained wind extrapolation can effectively bridge the observational gap above meteor radar altitudes, contributing to a better understanding of the physical mechanisms of Es layers and enhancing forecasting capabilities.
- 8 September 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 131 (9)
Physics‐Motivated Middle‐ to Low‐Latitude NmF2 Reconstruction Using Thermospheric ΣO/N 2 and Solar EUV Radiation
Traditional deep learning models for the ionosphere often rely on space weather indices as inputs, offering predictive capability but limited physical interpretability. We present SRON2NN, a physics‐motivated artificial neural network (ANN) that incorporates two geophysically motivated features: effective solar radiation (ESR) and the thermospheric column density ratio of atomic oxygen to molecular nitrogen (ΣO/N 2 ). Using a comprehensive data set comprising COSMIC‐1 radio occultation data, TIMED/Global Ultraviolet Imager ΣO/N 2 observations, and solar EUV measurements from 2008 to 2019, we find that SRON2NN consistently outperforms an index‐based ANN using only Dst, F10.7, time, and location across all testing conditions. The improvement is most evident during geomagnetic storms in mid‐latitude regions, especially in the Southern Hemisphere, where the contribution of ΣO/N 2 leads to an 8.20% reduction in root mean square error. Compared with adding vertical total electron content and IRI‐NmF2, our physics‐motivated features yield a more pronounced improvement in reconstructing NmF2. Cross‐model experiments suggest that incorporating these two features can consistently boost performance not only in ANN frameworks but also in tree‐based models, including random forests and XGBoost. Further analysis of four geomagnetic storm events shows that SRON2NN is generally consistent with storm‐time ΣO/N 2 depletion and the associated decrease in NmF2. Overall, these results suggest that incorporating ESR and ΣO/N 2 as physically motivated inputs can improve NmF2 reconstruction while enhancing interpretability.
- 8 September 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 131 (9)
Physics‐Motivated Middle‐ to Low‐Latitude NmF2 Reconstruction Using Thermospheric ΣO/N 2 and Solar EUV Radiation
Traditional deep learning models for the ionosphere often rely on space weather indices as inputs, offering predictive capability but limited physical interpretability. We present SRON2NN, a physics‐motivated artificial neural network (ANN) that incorporates two geophysically motivated features: effective solar radiation (ESR) and the thermospheric column density ratio of atomic oxygen to molecular nitrogen (ΣO/N 2 ). Using a comprehensive data set comprising COSMIC‐1 radio occultation data, TIMED/Global Ultraviolet Imager ΣO/N 2 observations, and solar EUV measurements from 2008 to 2019, we find that SRON2NN consistently outperforms an index‐based ANN using only Dst, F10.7, time, and location across all testing conditions. The improvement is most evident during geomagnetic storms in mid‐latitude regions, especially in the Southern Hemisphere, where the contribution of ΣO/N 2 leads to an 8.20% reduction in root mean square error. Compared with adding vertical total electron content and IRI‐NmF2, our physics‐motivated features yield a more pronounced improvement in reconstructing NmF2. Cross‐model experiments suggest that incorporating these two features can consistently boost performance not only in ANN frameworks but also in tree‐based models, including random forests and XGBoost. Further analysis of four geomagnetic storm events shows that SRON2NN is generally consistent with storm‐time ΣO/N 2 depletion and the associated decrease in NmF2. Overall, these results suggest that incorporating ESR and ΣO/N 2 as physically motivated inputs can improve NmF2 reconstruction while enhancing interpretability.
- Open Access
- 22 July 2026
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 53 (14)
Physics‐Embedded Deep Learning Reveals the Memory Effect of Ionospheric Electron Density in Response to Solar EUV Forcing
Precise characterization of the electron density profile is crucial for optimizing high frequency communications. Reconstruction models often assume an instantaneous ionospheric response to solar forcing or using time delays, neglecting the time‐integrated cumulative “memory effect” of plasma dynamics. Here, we develop the Electron Density Profile Reconstruction Neural Network (EDPRNN), a physics‐embedded framework that parameterizes and dynamically learns this memory representation from data. Using observations from three mid‐ and low‐latitude digisondes, we reveal that the learned memory exhibits distinct altitude and latitude dependence. Below the F1 layer, this data‐driven memory strongly correlates ( r ≈ 0.97) with theoretical photochemical time constants at all stations. EDPRNN also performs well on an independent solar maximum test set, with coefficient of determination ( R 2 ) values of 0.939, 0.905, and 0.841 at Mohe, Zuoling, and Fuke, respectively. These results indicate that EDPRNN provides an effective ionospheric nowcasting method and a novel data‐driven framework for identifying a physically consistent memory effect.
- 22 July 2026
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 53 (14)
Open AccessPhysics‐Embedded Deep Learning Reveals the Memory Effect of Ionospheric Electron Density in Response to Solar EUV Forcing
Precise characterization of the electron density profile is crucial for optimizing high frequency communications. Reconstruction models often assume an instantaneous ionospheric response to solar forcing or using time delays, neglecting the time‐integrated cumulative “memory effect” of plasma dynamics. Here, we develop the Electron Density Profile Reconstruction Neural Network (EDPRNN), a physics‐embedded framework that parameterizes and dynamically learns this memory representation from data. Using observations from three mid‐ and low‐latitude digisondes, we reveal that the learned memory exhibits distinct altitude and latitude dependence. Below the F1 layer, this data‐driven memory strongly correlates ( r ≈ 0.97) with theoretical photochemical time constants at all stations. EDPRNN also performs well on an independent solar maximum test set, with coefficient of determination ( R 2 ) values of 0.939, 0.905, and 0.841 at Mohe, Zuoling, and Fuke, respectively. These results indicate that EDPRNN provides an effective ionospheric nowcasting method and a novel data‐driven framework for identifying a physically consistent memory effect.
- Open Access
- 11 July 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 131 (7)
Simulation Study of HF Wave Propagation With EPB Effects Based on AI‐Assim Model
Ionospheric irregularities, particularly equatorial plasma bubbles (EPBs), can severely disrupt trans‐ionospheric high‐frequency (HF) wave propagation. This study employs a novel ionospheric electron density model for the low‐latitude region, referred to as AI‐Assim. Within this background, we construct a three‐dimensional Gaussian distribution model of EPBs at low latitudes. Calibrating it with airglow data from Hainan yields a Root Mean Square of ∼0.8 at the EPB center. A 3D ray‐tracing algorithm is then employed to compare HF wave propagation in the presence versus absence of EPB. At 10 and 12 MHz, radio rays traversing the EPB exhibit pronounced refraction and strong focusing, whereas waves in an undisturbed ionosphere penetrate the ionosphere. When radio waves propagate inside the EPB, the refractive effect significantly diminishes compared to that at the boundary. Further comparisons with the IRI‐2020 model reveal distinctly different ray paths: at 12 MHz, the IRI model yields partial reflection partial penetration, while the AI‐Assim model produces complete penetration. The simulation results based on AI‐Assim can provide scientific basis for OTHR detection in the presence of irregularities.
- 11 July 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 131 (7)
Open AccessSimulation Study of HF Wave Propagation With EPB Effects Based on AI‐Assim Model
Ionospheric irregularities, particularly equatorial plasma bubbles (EPBs), can severely disrupt trans‐ionospheric high‐frequency (HF) wave propagation. This study employs a novel ionospheric electron density model for the low‐latitude region, referred to as AI‐Assim. Within this background, we construct a three‐dimensional Gaussian distribution model of EPBs at low latitudes. Calibrating it with airglow data from Hainan yields a Root Mean Square of ∼0.8 at the EPB center. A 3D ray‐tracing algorithm is then employed to compare HF wave propagation in the presence versus absence of EPB. At 10 and 12 MHz, radio rays traversing the EPB exhibit pronounced refraction and strong focusing, whereas waves in an undisturbed ionosphere penetrate the ionosphere. When radio waves propagate inside the EPB, the refractive effect significantly diminishes compared to that at the boundary. Further comparisons with the IRI‐2020 model reveal distinctly different ray paths: at 12 MHz, the IRI model yields partial reflection partial penetration, while the AI‐Assim model produces complete penetration. The simulation results based on AI‐Assim can provide scientific basis for OTHR detection in the presence of irregularities.
- Open Access
- 11 March 2026
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 1000 (1) , 26
Particle-in-cell Simulations of Plasma Oscillations Excited by Dust Impacts
Dust-impact-induced plasma oscillations (dust ringing) detected in Saturn’s Enceladus plume provide a novel method for measuring electron density. However, since S. Ye et al. discovered such events, their underlying mechanisms remain unclear. In this study, linear instability analysis and particle-in-cell (PIC) simulations are employed to demonstrate that the beam–plasma instability mechanism is the more probable cause of dust-ringing events. Dust-ringing events were first observed in a pristine environment within the Enceladus plume, with a characteristic duration of approximately 1 ms under negative spacecraft potential. Our linear analysis demonstrates significant wave growth at a high beam density of 0.1 times the background electron density, occurring when the beam electron velocity surpasses the ambient electron thermal velocity by a factor of 4. Additionally, the PIC simulations confirm Langmuir wave excitation at kλ De ≈ 0–0.3, manifested as enhanced electric field oscillations and electron velocity scattering. These results suggest that when the plasma cloud generated by the dust impact maintains a relatively high density during initial diffusion and forms a rapid beam relative to the background plasma, it triggers beam–plasma instability, resulting in dust-ringing effects.
- 11 March 2026
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 1000 (1) , 26
Open AccessParticle-in-cell Simulations of Plasma Oscillations Excited by Dust Impacts
Dust-impact-induced plasma oscillations (dust ringing) detected in Saturn’s Enceladus plume provide a novel method for measuring electron density. However, since S. Ye et al. discovered such events, their underlying mechanisms remain unclear. In this study, linear instability analysis and particle-in-cell (PIC) simulations are employed to demonstrate that the beam–plasma instability mechanism is the more probable cause of dust-ringing events. Dust-ringing events were first observed in a pristine environment within the Enceladus plume, with a characteristic duration of approximately 1 ms under negative spacecraft potential. Our linear analysis demonstrates significant wave growth at a high beam density of 0.1 times the background electron density, occurring when the beam electron velocity surpasses the ambient electron thermal velocity by a factor of 4. Additionally, the PIC simulations confirm Langmuir wave excitation at kλ De ≈ 0–0.3, manifested as enhanced electric field oscillations and electron velocity scattering. These results suggest that when the plasma cloud generated by the dust impact maintains a relatively high density during initial diffusion and forms a rapid beam relative to the background plasma, it triggers beam–plasma instability, resulting in dust-ringing effects.
- Open Access
- 19 May 2025
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 985 (1) , 114
Quasithermal Noise in Magnetized Plasma: Theory and Simulations
Quasithermal noise (QTN) spectroscopy has been extensively used as an accurate tool to measure electron density and temperature in unmagnetized space plasmas. The observed spectrum depends on both antenna geometry and plasma kinetic properties. When considering the QTN signal of a magnetized plasma, the description becomes more complicated. We utilize particle-in-cell (PIC) simulations to study the QTN characteristics in magnetized plasmas based on Maxwellian and two-Maxwellian distributions. We compare these results with numerical computations, finding good agreement between the two. However, there are some minor differences. Specifically, when the angle between the wavevector (k) and the background magnetic field was 60°, clear harmonics appeared in the QTN signal from the numerical computations, whereas in the PIC simulations, these harmonics appeared at an angle of 80°. This discrepancy may be related to additional noise in the PIC simulation and our parameter limitations. In addition, Bernstein harmonics become more pronounced as the angle increases between k and the background magnetic field. These results demonstrate that PIC simulations can be used to more accurately characterize QTN signals in complex, magnetized systems.
- 19 May 2025
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 985 (1) , 114
Open AccessQuasithermal Noise in Magnetized Plasma: Theory and Simulations
Quasithermal noise (QTN) spectroscopy has been extensively used as an accurate tool to measure electron density and temperature in unmagnetized space plasmas. The observed spectrum depends on both antenna geometry and plasma kinetic properties. When considering the QTN signal of a magnetized plasma, the description becomes more complicated. We utilize particle-in-cell (PIC) simulations to study the QTN characteristics in magnetized plasmas based on Maxwellian and two-Maxwellian distributions. We compare these results with numerical computations, finding good agreement between the two. However, there are some minor differences. Specifically, when the angle between the wavevector (k) and the background magnetic field was 60°, clear harmonics appeared in the QTN signal from the numerical computations, whereas in the PIC simulations, these harmonics appeared at an angle of 80°. This discrepancy may be related to additional noise in the PIC simulation and our parameter limitations. In addition, Bernstein harmonics become more pronounced as the angle increases between k and the background magnetic field. These results demonstrate that PIC simulations can be used to more accurately characterize QTN signals in complex, magnetized systems.
- Open Access
- 27 August 2024
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 972 (1) , 90
Electrostatic Modeling of Dust Impact Signals Based on the Expanding Plasma Cloud
Signals generated by dust impacting spacecraft can be detected by electric field instruments. These signals have been simulated by numerous models. However, few models can accurately characterize the expansion of the plasma cloud generated by dust impact. The COMSOL model presented in this paper provides a way to understand the expansion properties of ions and electrons. The model can also be used to analyze the various expected waveforms of dust impact signals as a function of different parameters, such as the spacecraft voltage and the ambient plasma temperature. The results show that close to 50% of ions and electrons in the impact plasma cloud are collected by spacecraft at weak spacecraft potentials and that a fraction of the ions is still collected rather than all of them streaming away from the impact location at V SC > 0 V. The model also confirms that in the expanding plasma cloud, ions are in the form of plumes, while electrons diffuse in an isotropic manner.
- 27 August 2024
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 972 (1) , 90
Open AccessElectrostatic Modeling of Dust Impact Signals Based on the Expanding Plasma Cloud
Signals generated by dust impacting spacecraft can be detected by electric field instruments. These signals have been simulated by numerous models. However, few models can accurately characterize the expansion of the plasma cloud generated by dust impact. The COMSOL model presented in this paper provides a way to understand the expansion properties of ions and electrons. The model can also be used to analyze the various expected waveforms of dust impact signals as a function of different parameters, such as the spacecraft voltage and the ambient plasma temperature. The results show that close to 50% of ions and electrons in the impact plasma cloud are collected by spacecraft at weak spacecraft potentials and that a fraction of the ions is still collected rather than all of them streaming away from the impact location at V SC > 0 V. The model also confirms that in the expanding plasma cloud, ions are in the form of plumes, while electrons diffuse in an isotropic manner.
- 21 October 2023
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 128 (10)
Different Behaviors of Outer Radiation Belt keV and MeV Electrons During Two Sequential Geomagnetic Storms
Electron flux dropout is an extraordinary phenomenon in the outer radiation belt, which is characterized by a drastic depletion in trapped electron populations over a broad range of energies, pitch angles, and L‐shells. Using data from multiple satellites, we study a 5‐day dropout of MeV electrons and a several‐hour dropout of keV electrons in geostationary Earth orbit (GEO) during two sequential geomagnetic storms. Outward radial diffusion caused real loss of MeV electrons, while asymmetric and depressed magnetic fields led to adiabatic variations in keV electron fluxes. In addition to time scale differences, low‐energy keV, medium‐energy keV and high‐energy MeV electron fluxes behaved differently, involving an increasing trend, an initial decreasing trend followed by an increasing trend, and a decreasing trend, respectively, as observed by Radiation Belt Storm Probes (RBSPs) and low‐altitude satellites during the first storm. Substorm injections accompanied by inward radial diffusion initially produced an enhanced flux of electrons with energies up to ∼100 keV. Chorus waves generated in morning side of magnetosphere were responsible for scattering part of this seed population into the atmosphere. In turn, these enhanced but short‐lived chorus waves accelerated lower‐energy keV electrons to achieve a moderate energy level of hundreds of keV within one day. Additionally, the initial slow outward radial diffusion was associated with MeV electron loss, and there was insufficient time for acceleration by chorus waves, so only MeV electrons could maintain long‐term dropout. Competition between energy‐ and L‐shell‐dependent acceleration and loss mechanisms caused different keV and MeV electron behaviors.
- 21 October 2023
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 128 (10)
Different Behaviors of Outer Radiation Belt keV and MeV Electrons During Two Sequential Geomagnetic Storms
Electron flux dropout is an extraordinary phenomenon in the outer radiation belt, which is characterized by a drastic depletion in trapped electron populations over a broad range of energies, pitch angles, and L‐shells. Using data from multiple satellites, we study a 5‐day dropout of MeV electrons and a several‐hour dropout of keV electrons in geostationary Earth orbit (GEO) during two sequential geomagnetic storms. Outward radial diffusion caused real loss of MeV electrons, while asymmetric and depressed magnetic fields led to adiabatic variations in keV electron fluxes. In addition to time scale differences, low‐energy keV, medium‐energy keV and high‐energy MeV electron fluxes behaved differently, involving an increasing trend, an initial decreasing trend followed by an increasing trend, and a decreasing trend, respectively, as observed by Radiation Belt Storm Probes (RBSPs) and low‐altitude satellites during the first storm. Substorm injections accompanied by inward radial diffusion initially produced an enhanced flux of electrons with energies up to ∼100 keV. Chorus waves generated in morning side of magnetosphere were responsible for scattering part of this seed population into the atmosphere. In turn, these enhanced but short‐lived chorus waves accelerated lower‐energy keV electrons to achieve a moderate energy level of hundreds of keV within one day. Additionally, the initial slow outward radial diffusion was associated with MeV electron loss, and there was insufficient time for acceleration by chorus waves, so only MeV electrons could maintain long‐term dropout. Competition between energy‐ and L‐shell‐dependent acceleration and loss mechanisms caused different keV and MeV electron behaviors.
- Open Access
- 12 October 2020
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 125 (10)
Long‐Term Dropout of Relativistic Electrons in the Outer Radiation Belt During Two Sequential Geomagnetic Storms
On 31 January 2016, the flux of >2 MeV electrons observed by Geostationary Operational Environmental Satellite (GOES)‐13 dropped to the background level during a minor storm main phase (−48 nT). Then, a second storm (−53 nT) occurred on 2 February; during the 3 days after its main phase, the flux remained at background level. Using data from various instruments on the GOES, Polar Operational Environmental Satellites (POES), Radiation Belt Storm Probes (RBSP), Meteor‐M2, and Fengyun‐series spacecraft, we study this long‐term dropout of MeV electrons during two sequential storms of similar magnitude under lightly disturbed solar wind conditions. Observations from low‐altitude satellites show that the fluxes decreased first at higher L‐shells and then gradually propagated inward. Moreover, the fluxes were almost completely lost and dropped to the background level at L > 5, while the fluxes at 4 < L < 5 were partly lost, as observed by RBSP and low‐altitude satellites. Finally, observations show that on 5 February, only the fluxes at L > 5.5 recovered, while the fluxes at 4 < L < 5 did not return to the prestorm levels. These observations indicate that the loss and recovery processes developed first at higher L‐shells. Phase space density (PSD) analysis shows that radial outward diffusion was the main reason for the dropout at higher L‐shells. Regarding electron enhancement, stronger inward diffusion was accompanied by ultra‐low‐frequency (ULF) wave activities at higher L‐shells, and chorus waves observed at outer L‐shells provided conditions for relativistic electron flux recovery to the prestorm levels.
- 12 October 2020
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Space Physics
- Vol. 125 (10)
Open AccessLong‐Term Dropout of Relativistic Electrons in the Outer Radiation Belt During Two Sequential Geomagnetic Storms
On 31 January 2016, the flux of >2 MeV electrons observed by Geostationary Operational Environmental Satellite (GOES)‐13 dropped to the background level during a minor storm main phase (−48 nT). Then, a second storm (−53 nT) occurred on 2 February; during the 3 days after its main phase, the flux remained at background level. Using data from various instruments on the GOES, Polar Operational Environmental Satellites (POES), Radiation Belt Storm Probes (RBSP), Meteor‐M2, and Fengyun‐series spacecraft, we study this long‐term dropout of MeV electrons during two sequential storms of similar magnitude under lightly disturbed solar wind conditions. Observations from low‐altitude satellites show that the fluxes decreased first at higher L‐shells and then gradually propagated inward. Moreover, the fluxes were almost completely lost and dropped to the background level at L > 5, while the fluxes at 4 < L < 5 were partly lost, as observed by RBSP and low‐altitude satellites. Finally, observations show that on 5 February, only the fluxes at L > 5.5 recovered, while the fluxes at 4 < L < 5 did not return to the prestorm levels. These observations indicate that the loss and recovery processes developed first at higher L‐shells. Phase space density (PSD) analysis shows that radial outward diffusion was the main reason for the dropout at higher L‐shells. Regarding electron enhancement, stronger inward diffusion was accompanied by ultra‐low‐frequency (ULF) wave activities at higher L‐shells, and chorus waves observed at outer L‐shells provided conditions for relativistic electron flux recovery to the prestorm levels.