We report on observations of a set of magnetic loops above a region with late-phase flux emergence taken by IRIS, Hinode, and SDO. The loop system consists of many transition-region loop threads that are 5″-12″ in length and ∼0.″5 in width and coronal loops with similar length and ∼2″ width. In the middle of the loop system, there is a clear systematic blueshift of about 10 km s^-1 in the transition region that is consistent with a flux-emerging picture, while a redshift of about 10 km s^-1 in the corona is observed. The nonthermal velocity of the loop system is smaller than that of the surrounding region in the transition region but is comparable that in the corona. The electron density of a transition-region loop is also measured and found to be about 5 × 10¹⁰ cm^-3, a magnitude larger than that in the coronal loops. In agreement with imaging data, the temperature profiles derived from the differential emission measurement technique confirm that some of the loops have been heated to corona level. The expansion of the loops leads to interactions between the loops themselves and with the ambient field, which can drive magnetic reconnection evidenced by multiple intense brightenings, including transition-region explosive events and IRIS bombs in the footpoint region associated with the moving polarity. A set of quasi-periodic brightenings with a period of about 130 s is found at the loop apex. Our observations indicate that the flux emergence in its late phase is much different from that at the early stage. While the observed transition region is dominated by emerging flux, these emerging loops could be heated to corona level, and the heating (if via nonthermal processes) most likely takes place only after they reach the transition region or lower corona.

Emerging flux regions are the locations where a variety of sporadic local heating events are observed. With the aim of understanding such events, we analyze the observational data of emerging AR 12401 obtained by Hinode/SOT and IRIS. Out of the 29 Ca bright points (BPs), seven are situated at the mixed-polarity photospheric background, above which the IRIS UV spectra are strongly enhanced and broadened, indicating the bi-directional jets from low altitude magnetic reconnection. Another 10 BPs are found in unipolar regions at the edges of the emerging flux region. Their UV spectra are in general weak but consistently redshifted at the speeds up to 40 km/s. These observational results support the physical picture that the heating in the region center are due to magnetic reconnection at flux cancellation sites (Ellerman bombs and UV bursts), whereas the peripheral events are due to shocks caused by supersonic downflows along the rising arch filament system.

Coronal loops are building blocks of solar active regions (ARs). However, their formation is not well understood. Here we present direct observational evidence for the formation of coronal loops through magnetic reconnection as new magnetic fluxes emerge to the solar atmosphere. Observations in the EUV passbands of SDO/AIA clearly show the newly formed loops following magnetic reconnection within a vertical current sheet. Formation of the loops is also seen in the Hα images taken by NVST. The SDO/HMI observations show that a positive-polarity flux concentration moves toward a negative-polarity one with a speed of ~0.5 km s^-1 before the apparent formation of coronal loops. During the formation of coronal loops, we found signatures of flux cancellation and subsequent enhancement of the transverse field between the two polarities. We have reconstructed the three-dimensional magnetic field structure through a magnetohydrostatic model, which shows field lines consistent with the loops in AIA images. Numerous bright blobs with a width of ~1.5 Mm appear intermittently in the current sheet and move upward with apparent velocities of ~80 km s^-1. We have also identified plasma blobs moving to the footpoints of the newly formed large loops, with apparent velocities ranging from 30 to 50 km s^-1. A differential emission measure analysis shows that the temperature, emission measure and density of the bright blobs are 2.5-3.5 MK, 1.1-2.3×10²⁸ cm^-5 and 7.9-16.4×10⁹ cm^-3, respectively. Power spectral analysis of these blobs indicates that the magnetic reconnection is inconsistent with the turbulent reconnection scenario.

White-light flares (WLFs), first observed in 1859, refer to a type of solar flare showing an obvious enhancement of the visible continuum emission. This type of enhancement often occurs in most energetic flares, and is usually interpreted as a consequence of efficient heating in the lower solar atmosphere through nonthermal electrons propagating downward from the energy release site in the corona. However, this coronal-reconnection model has difficulty in explaining the recently discovered small WLFs. Here we report a C2.3 WLF, which is associated with several observational phenomena: a fast decrease in opposite-polarity photospheric magnetic fluxes, the disappearance of two adjacent pores, significant heating of the lower chromosphere, a negligible increase of the hard X-ray flux, and an associated U-shaped magnetic field configuration. All these suggest that this WLF is powered by magnetic reconnection in the lower part of the solar atmosphere rather than by reconnection higher up in the corona.

We carry out radiation magnetohydrodynamics (RMHD) simulations for the sunspot formation. It is thought that the magnetic field is created deep in the convection zone, and thermal convection transports the magnetic flux to the surface of the Sun. Due to the significant spatial and temporal differences between the deep convection zone and the near-surface region, calculations in these layers have been separated. We developed the new RMHD code that can simultaneously deal with low and high Mach number situations in these regions. With the new code, we carry out a series of RMHD simulations of the sunspot formation in an unprecedentedly deep domain covering the whole depth of the convection zone. Large-scale convection in the deep layer of the Sun distorts the flux tube and creates sunspots at the solar surface. We find several possible influences of the bottom boundary condition on the simulation results in the previous shallow simulations. We also find that the thermal convection occasionally creates delta-type sunspots and complicated magnetic features at the surface. These results show the importance of the deep convection structure for investigating the sunspot formation at the surface.

We develop a method to predict the horizontal velocity field from the radiation intensity and vertical velocity on the solar surface using neural networks. We can observe thermal convection on the solar photosphere as a mottled pattern called granulation. The thermal convection is related to various phenomena such as the dynamo and the generation of the wave. While the line-of-sight (LOS) velocity can be observed by the Doppler effect, we cannot observe the velocity perpendicular to the LOS. There are methods to estimate the horizontal flow on the solar surface by tracking the time variation of the observation images called local correlation tracking. However, these require at least two or more images. On the other hand, many magnetohydrodynamic (MHD) simulations are carried out and reproduce the observations well. By using simulations, we can obtain physical quantities that cannot be observed. In this study, we use the results of our own MHD simulations to train neural networks, and this network predicts the unobservable horizontal velocity field from the observable intensity and vertical velocity field. By using only the convolution layer, images of any size can be applied, and the prediction is fast. The correlation coefficient between the simulation and the prediction is about 0.65, and this method can be applied to observations.

Solar coronal cavities are dark structures with a rarefied density compared with surrounding streamers. They are often observed as a component of the classic three-part structure of a coronal mass ejections (CME), quiescent cavities are observed mostly in the polar crown regions and may be long-lived. Some of the quiescent cavities may finally erupt as a CME. Cavities possess a characteristic “lagomorphic” structure, observed in linear polarisation observations, which can be explained with the flux rope model. Cavities are hotter than surrounding streamer, and the temperature remains stable.We present statistical analysis of all cavities observed by Coronal Multichannel Polarimeter (CoMP). These observations reveal that all cavities observed over a long time possess “lagomorphic” structure, therefore flux rope model is appropriate for polar-crown prominence cavities. Based on forward modelling we show that visibility of the structure in the linear polarisation depends on the orientation of the structure. We also present Differential Emission Measure (DEM) analysis of previously quiescent cavities during the eruption.  

In ultraviolet (UV) spectropolarimetric observations of the solar corona, the existence of magnetic field, solar wind velocity and temperature anisotropies modify the linear polarization associated with resonant scattering. In this paper, we forward model expected signals in the Lyman-α line in order to establish the different roles played by these physical effects. Unlike previous empirical models or global models, which present blended results of the above physical effects, the analytic model adopted here can be adjusted to test the roles of different effects separately. We find that the impact of all three effects is most evident in the rotation of the linear polarization direction. In particular: 1) a magnetic field on the order of several Gauss modifies the linear polarization at low coronal heights, rotating the linear polarization direction either clockwise or counter-clockwise when the angle between the magnetic field and the local vertical is greater or less than the van Vleck angle, which is consistent with the result of Zhao et al. (2019, Paper I hereafter). 2) Solar wind velocity, which increases with height, has a significant effect at higher coronal heights, rotating the linear polarization direction in an opposite fashion to the magnetic effect. 3) Temperature anisotropies are most significant at lower heights where the magnetic field departs from the radial, and follow the pattern of the magnetic effect, opposite to the Doppler dimming. The fact that the three effects operate differently in distinct spatial regimes opens up the possibility for using linear polarization measurements in UV lines to diagnose these important physical characteristics of the solar corona.

Magnetoseismology, a technique of magnetic field diagnostics based on observations of magnetohydrodynamic (MHD) waves, has been widely used to estimate the field strengths of oscillating structures in the solar corona. However, previously magnetoseismology was mostly applied to occasionally occurring oscillation events, providing an estimate of only the average field strength or one-dimensional distribution of field strength along an oscillating structure. This restriction could be eliminated if we apply magnetoseismology to the pervasive propagating transverse MHD waves discovered with the Coronal Multi-channel Polarimeter (CoMP). Using several CoMP observations of the Fe XIII 1074.7 nm and 1079.8 nm spectral lines, we obtained maps of the plasma density and wave phase speed in the corona, which allow us to map both the strength and direction of the coronal magnetic field in the plane of sky. We also examined distributions of the electron density and magnetic field strength, and compared their variations with height in the quiet Sun and active regions. Such measurements could provide critical information to advance our understanding of the Sun's magnetism and the magnetic coupling of the whole solar atmosphere.

In this work we will describe Hinode/EIS measurements of coronal magnetic fields in active regions and before and during flares using the newly developed Magnetically-Induced Transition (MIT) technique. We will compare measurements to magnetic field extrapolations, as well as discuss the importance of these measurements for flare forecasting.

Measurements from the Van Allen Probes mission clearly demonstrated that the radiation belts cannot be considered as a bulk population above approximately electron rest mass. Ultra-relativistic electrons (~>4Mev) form a new population that shows a very different morphology (e.g. very narrow remnant belts) and slow but sporadic acceleration. We show that acceleration to multi-MeV energies can not only result of a two-step processes consisting of local heating and radial diffusion but occurs locally due to energy diffusion by whistler mode waves. Local heating appears to be able to transport electrons in energy space from 100s of keV all the way to ultra-relativistic energies (>7MeV). Acceleration to such high energies occurs only for the conditions when cold plasma in the trough region is extremely depleted down to the values typical for the plasma sheet. There is also a clear difference between the loss mechanisms at MeV and multi MeV energies. The difference between the loss mechanisms at MeV and multi-MeV energies is due to EMIC waves that can very efficiently scatter ultra-relativistic electrons, but leave MeV electrons unaffected. We also present how the new understanding gained from the Van Allen Probes mission can be used to produce the most accurate data assimilative forecast. Under the recently funded EU Horizon 2020 Project Prediction of Adverse effects of Geomagnetic storms and Energetic Radiation (PAGER) we will study how ensemble forecasting from the Sun can produce long-term probabilistic forecasts of the radiation environment in the inner magnetosphere.

Electromagnetic ion cyclotron (EMIC) waves have long been known to drive the scattering loss of relativistic electrons from the radiation belts into the Earth's upper atmosphere. In recent years, however, there has been growing evidence to suggest that EMIC can cause the loss of not only relativistic electrons, but also sub-relativistic electrons down to energies of only a few hundred keV. At these lower energies, EMIC waves have access to much larger trapped electron populations, and thus have the potential to drive sufficient electron precipitation to be a significant source of ionisation in the upper atmosphere.
In this presentation we first describe an apparent contradiction between studies of trapped and precipitation electron fluxes in the presence of EMIC wave activity. We demonstrate a simple resolution to this contradiction, and show that that a purely relativistic trapped flux response to EMIC wave activity is not necessarily mutually exclusive with strong sub-MeV electron precipitation measurements.
We also present results from the Sodankylä ion-chemistry model, which we use to investigate the impact of EMIC-driven electron precipitation on atmospheric chemistry. We show that EMIC-driven electron precipitation can drive significant ionisation in the Earth's atmosphere above 40 km, leading to the loss of mesospheric ozone. We compare these results with existing geomagnetic activity proxies used as drivers of modern coupled-climate models, suggesting that EMIC-driven electron precipitation is not currently being accounted for in these models.

Precipitating electrons and protons from the aurora and large solar eruptions are known sources of nitric oxide in the high-latitude mesosphere and lower thermosphere. This NO formation by precipitating energetic particles initiates a chain of chemical-dynamical coupling processes significantly affecting polar stratospheric ozone, and impacting atmospheric dynamics possibly down to tropospheric weather systems. Consequently, this “geomagnetic forcing” is increasingly studied in global chemistry-climate model experiments to quantify the natural forcing of the climate system. There is emerging evidence that electrons accelerated in the radiation belts contribute significantly to this geomagnetic forcing; however, their frequency, intensity and energy range are still uncertain.  In this presentation, we will provide a summary about the state-of-the art of the geomagnetic forcing with a focus on the emerging role of precipitating magnetospheric electrons.

Auroral field-aligned currents (FACs), which are thought to be highly related to particle precipitations, are an important transport mechanism for energy and momentum between the magnetosphere and ionosphere. In this presentation, we derived FACs from DMSP satellites. The magnetic latitude versus local time distribution of FACs from DMSP shows comparable dependences with previous findings on the intensity and orientation of interplanetary magnetic field (IMF) By and Bz components, which also confirms the reliability of the DMSP FAC data set. With simultaneous measurements of precipitating particles from DMSP, we further investigate the relation between large-scale FACs and precipitating particle energy flux. Our result shows that precipitation electron and ion fluxes both increase in magnitude and extend to lower latitude for enhanced southward IMF Bz, which is similar to the behavior of FACs. Under weak northward and southward Bz conditions, the location of the R2 current maxima, at both dusk and dawn sides and both hemispheres, are found to be close to the maxima of the particle energy fluxes; while for the same IMF conditions, R1 currents are displaced further to the respective particle flux peaks. Largest displacement (about 3.5⁰) is found between the downward R1 current and ion flux peak at the dawn side. Our results suggest that there exist systematic differences in the peak locations of electron/ion precipitation and large-scale upward/downward FACs, with FAC peaks enclosing the particle energy flux peaks in auroral band at both dusk and dawn sides. Our comparisons also found that particle precipitation at dawn and dusk and in both hemispheres maximizes near the mean R2 current peaks. The particle precipitation maxima closer to the R1 current peaks are lower in magnitude, which is opposite to the known feature that R1 currents are on average stronger than R2 currents.

Earth's ionosphere represents a complex and dynamic region characterized by increased concentration of charged particles. Changes in ionospheric density can affect the propagation of electromagnetic signals thus disrupting navigation and positioning. The existing models of ionospheric density rarely meet the accuracy requirements due to being either only climatological (physical models), or using the time and spatial averaging (empirical models). Here we present a continuous empirical three dimensional (3D) model of electron density at heights 130-900 km. Since the ionosphere is a data rich environment it is essential to use all of the collected observations. We use the radio occultation data from various spacecraft, together with in situ data by CHAMP and CNOFS missions and observations from GRACE-KBR, for over 18 years of data. We discuss the application of deep learning to efficiently handle the entire dataset comprising billions of data points. In order to analyze, which features result in the best model performance, we employ a number of ML-based feature selection techniques and discuss the optimal combination of inputs, and their physical meaning. The resulting model gives accurate predictions of electron density in the Earth's ionosphere and yields >90% correlation on the validation data. The model has a wide range of applications for the scientific purposes, space weather monitoring and industrial applications such as positioning and navigation.

We conduct observational and modeling studies of thermospheric composition responses to weak geomagnetic activity (non-geomagnetic storms). We found that the thermospheric O and N₂ column density ratio (∑O/N₂) in part of the Northern Hemisphere measured by Global-scale Observations of the Limb and Disk (GOLD) exhibited large and long-lived depletions during weak geomagnetic activity in May and June 2019. The depletions reached 30% of quiet time values, extended equatorward to 10°N, and lasted more than 10 hours. Furthermore, numerical simulation results are similar to these observations and indicate that the ∑O/N₂ depletions were pushed westward by zonal winds. The ∑O/N₂ evolution during weak geomagnetic activity suggests that the formation mechanism of the ∑O/N₂ depletions is similar to that during a geomagnetic storm. The effects of weak geomagnetic activity are often ignored, but, in fact, are important for understanding thermosphere neutral composition variability and hence the state of the thermosphere-ionosphere system.

Observations by the GOLD mission are able to follow spatial and temporal changes in thermospheric composition, density and temperature and in ionospheric structure and peak density, providing fundamental information for understanding the Thermosphere-Ionosphere system. GOLD images the Earth from geostationary orbit at 47.5° W longitude, observing on a ≤30-minute cadence from 06:10 to 0:40 UT (03:00-21:30 LT at the satellite). Observations performed include limb scans; stellar occultations; and images the sunlit and nightside disk.  These data provide simultaneous, synoptic imaging of the composition and temperature near 160 km for the first time, as well as follow the temporal development of the nighttime ionosphere over the Atlantic and South America every evening. Recent observations of geomagnetic storms and solar eclipses have produced some surprising results. These observations are providing significant tests of current models, tests that may lead to advances in the capabilities of global-scale models of the thermosphere and ionosphere system. Observations also clearly show the presence of waves in the thermosphere, as well as indications at multiple spatial of the influence of waves and tides propagating from lower altitudes. Examples of these observations, as well as their potential for advancing models of the thermosphere-ionosphere system, will be presented and discussed.

This study investigates the underlying physics of equatorial plasma bubbles (EPBs) on 11 December 2019, under solar minimum conditions. The Global‐scale Observations of the Limb and Disk (GOLD) ultraviolet nightglow images exhibit a periodic distribution of reduced emissions related to EPBs. Remarkably, FORMOSAT‐7/COSMIC‐2 (F7/C2) observes a significant altitudinal difference of ~45 km in the bottomside ionosphere between two nearly collocated electron density profiles before the onset of EPBs, indicating the presence of an upwelling. Distinct ionospheric perturbations are also observed in F7/C2 and ground‐based Global Positioning System observations, suggesting that gravity waves may contribute to the upwelling. Simulations with SAMI3/ESF are further carried out to evaluate the upwelling growth and pre‐reversal enhancement (PRE) effect on EPB development. Simulations reveal that the crests of upwellings show a localized uplift of ~50 km, and EPBs only develop from the crest of upwellings. The uplift altitude of upwellings is comparable to the F7/C2 observations and the post‐sunset rise in moderate solar conditions. The polarization electric field (E_p) developed within the upwellings can drive vertical E_p × B drifts of ~32–35 m/s, which are comparable to the PRE vertical E × B drifts. We find that the PRE alone cannot drive EPBs without upwelling growth, but it can facilitate the upwelling growth. Observations and simulations allow us to conclude that upwelling growth could play a vital role in the formation of EPBs.

Equatorial ionosphere can react to substorm events via direct penetrating electric field.   In the past, the thermosphere and ionosphere model driven by the high empirical ion convection models lacked the dynamic high latitude input to properly simulate the penetrating electric field and equatorial response.   The recently developed magnetospheric and ionospheric coupling model GTR (GAMERA-TIEGCM-RCM) aims to resolve the issue.   We will simulate recent substorm events to examine the penetrating electric field effect.    Additionally, we will use the observations from recent missions such as GOLD, COSMIC 2, and ICON to validate the simulation results.  These three missions provide unprecedented coverage of the equatorial ionosphere including ion density profiles, UV images, ion drift, and neutral winds.   The ion drift and neutral winds are particularly important for understanding the equatorial dynamo.