We characterize and catalog 30 solar eruptive events observed by the Fermi Large Area Telescope (LAT) having late-phase >100 MeV γ -ray emission (LPGRE), identified 30 yr ago in what were called long-duration gamma-ray flares. We show that LPGRE is temporally and spectrally distinct from impulsive phase emission in these events. The spectra are consistent with the decay of pions produced by >300 MeV protons and are not consistent with primary electron bremsstrahlung. Impulsive >100 keV X-ray emission was observed in all 27 LPGRE events where observations were made. All but two of the LPGRE events were accompanied by a fast and broad coronal mass ejection (CME). The LPGRE start times range from CME onset to 2 hr later. Their durations range from ∼0.1 to 20 hr and appear to be correlated with durations of >100 MeV solar energetic particle (SEP) proton events. The power-law spectral indices of the >300 MeV protons producing LPGRE range from ∼2.5 to 6.5 and vary during some events. Combined γ -ray line and LAT measurements indicate that LPGRE proton spectra are steeper above 300 MeV than they are below 300 MeV. The number of LPGRE protons >500 MeV is typically about 10× the number in the impulsive phase of the solar eruptive event and ranges in nine events from ∼0.01× to 0.5× the number in the accompanying SEP event, with large systematic uncertainty. What appears to be late-phase electron bremsstrahlung with energies up to ∼10 MeV was observed in one LPGRE event. We discuss how current models of LPGRE may explain these characteristics.

Context. The observation of > 100 MeV gamma-rays in the minutes to hours following solar flares suggests that high-energy particles interacting in the solar atmosphere can be stored and/or accelerated for long time periods. The occasions when gamma-rays are detected even when the solar eruptions occurred beyond the solar limb as viewed from Earth provide favorable viewing conditions for studying the role of coronal shocks driven by coronal mass ejections (CMEs) in the acceleration of these particles. Aims. In this paper, we investigate the spatial and temporal evolution of the coronal shocks inferred from stereoscopic observations of behind-the-limb flares to determine if they could be the source of the particles producing the gamma-rays. Methods. We analyzed the CMEs and early formation of coronal shocks associated with gamma-ray events measured by the Fermi-Large Area Telescope (LAT) from three eruptions behind the solar limb as viewed from Earth on 2013 Oct. 11, 2014 Jan. 06 and Sep. 01. We used a 3D triangulation technique, based on remote-sensing observations to model the expansion of the CME shocks from above the solar surface to the upper corona. Coupling the expansion model to various models of the coronal magnetic field allowed us to derive the time-dependent distribution of shock Mach numbers and the magnetic connection of particles produced by the shock to the solar surface visible from Earth. Results. The reconstructed shock fronts for the three events became magnetically connected to the visible solar surface after the start of the flare and just before the onset of the > 100 MeV gamma-ray emission. The shock surface at these connections also exhibited supercritical Mach numbers required for significant particle energization. The strongest gamma-ray emissions occurred when the flanks of the shocks were connected in a quasi-perpendicular geometry to the field lines reaching the visible surface. Multipoint, in situ, measurements of solar energetic particles (SEPs) were consistent with the production of these SEPs by the same shock processes responsible for the gamma-rays. The fluxes of protons in space and at the Sun were highest for the 2014 Sep. 01, which had the fastest moving shock. Conclusions. This study provides further evidence that high-energy protons producing time-extended high-energy gamma-ray emission likely have the same CME-shock origin as solar energetic particles measured in interplanetary space.

ABSTRACT The 3He abundance in impulsive solar energetic particle (SEP) events is enhanced up to several orders of magnitude compared to its photospheric value of [3He]/[4He] = 1–3 × 10−4. Interplanetary magnetic field and timing observations suggest that these events are related to solar flares. Observations of 3He in flare-accelerated ions would clarify the relationship between these two phenomena. Energetic 3He interactions in the solar atmosphere produce gamma-ray nuclear-deexcitation lines, both lines that are also produced by protons and α particles and lines that are essentially unique to 3He. Gamma-ray spectroscopy can, therefore, reveal enhanced levels of accelerated 3He. In this paper, we identify all significant deexcitation lines produced by 3He interactions in the solar atmosphere. We evaluate their production cross sections and incorporate them into our nuclear deexcitation-line code. We find that enhanced 3He can affect the entire gamma-ray spectrum. We identify gamma-ray line features for which the yield ratios depend dramatically on the 3He abundance. We determine the accelerated 3He/α ratio by comparing these ratios with flux ratios measured previously from the gamma-ray spectrum obtained by summing the 19 strongest flares observed with the Solar Maximum Mission Gamma-Ray Spectrometer. All six flux ratios investigated show enhanced 3He, confirming earlier suggestions. The 3He/α weighted mean of these new measurements ranges from 0.05 to 0.3 (depending on the assumed accelerated α/proton ratio) and has a <1 × 10−3 probability of being consistent with the photospheric value. With the improved code, we can now exploit the full potential of gamma-ray spectroscopy to establish the relationship between flare-accelerated ions and 3He-rich SEPs.

Measurements of the 0.511MeV positron-annihilation line from solar flares are used to explore the flare process in general and ion acceleration in particular. In flares, positrons are produced primarily by the decay of radioactive positron-emitting isotopes resulting from nuclear interactions of flare-accelerated ions with ambient solar material. Kozlovsky et al. provided ion-energy-dependent production cross sections for 67 positron emitters evaluated from their threshold energies (some <1 MeV nucleon(-1)) to a GeV nucleon(-1), incorporating them into a computer code for calculating positron-emitter production. Adequate cross-section measurements were available for proton reactions, but not for alpha-particle reactions where only crude estimates were possible. Here we re-evaluate the alpha-particle cross sections using new measurements and nuclear reaction codes. In typical large gamma-ray line flares, proton reactions dominate positron production, but alpha-particle reactions will dominate for steeper accelerated-ion spectra because of their relatively low threshold energies. With the accelerated-He-3 reactions added previously, the code is now reliable for calculating positron production from any distribution of accelerated-ion energies, not just those of typical flares. We have made the code available in the online version of the Journal. We investigate which reactions, projectiles, and ion energies contribute to positron production. We calculate ratios of the annihilation-line fluence to fluences of other gamma-ray lines. Such ratios can be used in interpreting flare data and in determining which nuclear radiation is most sensitive for revealing acceleration of low-energy ions at the Sun.

Major solar eruptive events (SEEs), consisting of both a large flare and a near simultaneous large fast coronal mass ejection (CME), are the most powerful explosions and also the most powerful and energetic particle accelerators in the solar system, producing solar energetic particles (SEPs) up to tens of GeV for ions and hundreds of MeV for electrons. The intense fluxes of escaping SEPs are a major hazard for humans in space and for spacecraft. Furthermore, the solar plasma ejected at high speed in the fast CME completely restructures the interplanetary medium (IPM) - major SEEs therefore produce the most extreme space weather in geospace, the interplanetary medium, and at other planets. Thus, understanding the flare/CME energy release process(es) and the related particle acceleration processes are major goals in Heliophysics. To make the next major breakthroughs, we propose a new mission concept, SEE 2020, a single spacecraft with a complement of advanced new instruments that focus directly on the coronal energy release and particle acceleration sites, and provide the detailed diagnostics of the magnetic fields, plasmas, mass motions, and energetic particles required to understand the fundamental physical processes involved.

We have evaluated the energetics of 38 solar eruptive events observed by a variety of spacecraft instruments between February 2002 and December 2006, as accurately as the observations allow. The measured energetic components include: (1) the radiated energy in the GOES 1 { 8 A band; (2) the total energy radiated from the soft X-ray (SXR) emitting plasma; (3) the peak energy in the SXR-emitting plasma; (4) the bolometric radiated energy over the full duration of the event; (5) the energy in are-accelerated electrons above 20 keV and in ions above 1 MeV; (6) the kinetic and potential energies of the coronal mass ejection (CME); (7) the energy in solar energetic particles (SEPs); and (8) the amount of free (nonpotential) magnetic energy estimated to be available in the pertinent active region. Major conclusions include: (1) the energy radiated by the SXR-emitting plasma exceeds, by about half an order of magnitude, the peak energy content of the thermal plasma that produces this radiation; (2) the energy content in are-accelerated electrons and ions is sufficient to supply the bolometric energy radiated across all wavelengths throughout the event; (3) the energy contents of are-accelerated electrons and ions are comparable; (4) the energy in SEPs is typically a few percent of the CME kinetic energy (measured in the rest frame of the solar wind); and (5) the available magnetic energy is sufficient to power the CME, the are-accelerated particles, and the hot thermal plasma.

We report on the detection of high-energy gamma-ray emission from the Moon during the first 24 months of observations by the Fermi Large Area Telescope (LAT). This emission comes from particle cascades produced by cosmic-ray (CR) nuclei and electrons interacting with the lunar surface. The differential spectrum of the Moon is soft and can be described as a log-parabolic function with an effective cutoff at 2-3 GeV, while the average integral flux measured with the LAT from the beginning of observations in 2008 August to the end of 2010 August is F(> 100 MeV) = (1.04 +/- 0.01 [statistical error] +/- 0.1 [systematic error]) x 10(-6) cm(-2) s(-1). This flux is about a factor 2-3 higher than that observed between 1991 and 1994 by the EGRET experiment on board the Compton Gamma Ray Observatory, F(> 100 MeV) approximate to 5 x 10(-7) cm-2 s-1, when solar activity was relatively high. The higher gamma-ray flux measured by Fermi is consistent with the deep solar minimum conditions during the first 24 months of the mission, which reduced effects of heliospheric modulation, and thus increased the heliospheric flux of Galactic CRs. A detailed comparison of the light curve with McMurdo Neutron Monitor rates suggests a correlation of the trends. The Moon and the Sun are so far the only known bright emitters of gamma-rays with fast celestial motion. Their paths across the sky are projected onto the Galactic center and high Galactic latitudes as well as onto other areas crowded with high-energy gamma-ray sources. Analysis of the lunar and solar emission may thus be important for studies of weak and transient sources near the ecliptic.

Neutron detectors on spacecraft in the inner heliosphere can observe the low-energy (<30 MeV) solar-flare neutrons that are not easily observable at Earth because they are lost to decay during transit. We present calculations of low-energy neutron production using a computer code incorporating updated neutron-production cross sections for the proton and α-particle reactions with heavier elements at all ion energies, especially at low energies (Eion < 10 MeV nucleon−1) most important for producing low-energy neutrons from these reactions. We calculate escaping-neutron spectra and neutron-capture line yields from ions propagating in a magnetic loop with various kinetic-energy spectra. This study provides the basis for planning inner-heliospheric missions having a low-energy neutron detector. The MESSENGER spacecraft orbiting Mercury has such a detector. We conclude that a full understanding of ion acceleration, transport, and interaction at the Sun requires observation of both neutrons and gamma rays with detectors of comparable sensitivity. We find that the neutron-capture line fluence at 1 AU is comparable to the 1–10 MeV neutron fluence at 0.5 AU, and therefore as effective for revealing low-energy ion acceleration. However, as the distance from the Sun to the neutron detector decreases, the tremendous increase of the low-energy neutron flux allows exploration of ion acceleration in weak flares not previously observable and may reveal acceleration at other sites not previously detected where low-energy neutrons could be the only high-energy signature of ion acceleration. Also, a measurement of the low-energy neutron spectrum will provide important information about the accelerated-ion spectrum that is not available from the capture line fluence measurement alone.

We report on the detection of high-energy γ-ray emission from the Moon during the first 24 months of observations by the Fermi Large Area Telescope (LAT). This emission comes from particle cascades produced by cosmic-ray (CR) nuclei and electrons interacting with the lunar surface. The differential spectrum of the Moon is soft and can be described as a log-parabolic function with an effective cutoff at 2–3 GeV, while the average integral flux measured with the LAT from the beginning of observations in 2008 August to the end of 2010 August is F(> cm−2 s−1. This flux is about a factor 2–3 higher than that observed between 1991 and 1994 by the EGRET experiment on board the Compton Gamma Ray Observatory, F(>100 MeV) ≈ 5 × 10−7 cm−2 s−1, when solar activity was relatively high. The higher γ-ray flux measured by Fermi is consistent with the deep solar minimum conditions during the first 24 months of the mission, which reduced effects of heliospheric modulation, and thus increased the heliospheric flux of Galactic CRs. A detailed comparison of the light curve with McMurdo Neutron Monitor rates suggests a correlation of the trends. The Moon and the Sun are so far the only known bright emitters of γ-rays with fast celestial motion. Their paths across the sky are projected onto the Galactic center and high Galactic latitudes as well as onto other areas crowded with high-energy γ-ray sources. Analysis of the lunar and solar emission may thus be important for studies of weak and transient sources near the ecliptic.

Due to an error at the publisher, the times given for the major tick marks in the X-axis in Figure 1 of the published article are incorrect. The correctly labeled times should be "00:52:00," "00:54:00," ... , and "01:04:00." The correct version of Figure 1 and its caption is shown below. IOP Publishing sincerely regrets this error.

We have evaluated the energetics of 38 solar eruptive events observed by a variety of spacecraft instruments between 2002 February and 2006 December, as accurately as the observations allow. The measured energetic components include: (1) the radiated energy in the Geostationary Operational Environmental Satellite 1-8 angstrom band, (2) the total energy radiated from the soft X-ray (SXR) emitting plasma, (3) the peak energy in the SXR-emitting plasma, (4) the bolometric radiated energy over the full duration of the event, (5) the energy in flare-accelerated electrons above 20 keV and in flare-accelerated ions above 1MeV, (6) the kinetic and potential energies of the coronal mass ejection (CME), (7) the energy in solar energetic particles (SEPs) observed in interplanetary space, and (8) the amount of free (non-potential) magnetic energy estimated to be available in the pertinent active region. Major conclusions include: (1) the energy radiated by the SXR-emitting plasma exceeds, by about half an order of magnitude, the peak energy content of the thermal plasma that produces this radiation; (2) the energy content in flare-accelerated electrons and ions is sufficient to supply the bolometric energy radiated across all wavelengths throughout the event; (3) the energy contents of flare-accelerated electrons and ions are comparable; (4) the energy in SEPs is typically a few percent of the CME kinetic energy (measured in the rest frame of the solar wind); and (5) the available magnetic energy is sufficient to power the CME, the flare-accelerated particles, and the hot thermal plasma.

The Geostationary Operational Environmental Satellite (GOES) M2-class solar flare, SOL2010-06-12T00: 57, was modest in many respects yet exhibited remarkable acceleration of energetic particles. The flare produced an similar to 50 s impulsive burst of hard X-and gamma-ray emission up to at least 400 MeV observed by the Fermi Gamma-ray Burst Monitor and Large Area Telescope experiments. The remarkably similar hard X-ray and high-energy gamma-ray time profiles suggest that most of the particles were accelerated to energies greater than or similar to 300 MeV with a delay of similar to 10 s from mildly relativistic electrons, but some reached these energies in as little as similar to 3 s. The gamma-ray line fluence from this flare was about 10 times higher than that typically observed from this modest GOES class of X-ray flare. There is no evidence for time-extended >100 MeV emission as has been found for other flares with high-energy gamma-rays.

We report the detection of high-energy gamma rays from the quiescent Sun with the Large Area Telescope (LAT) on board the Fermi Gamma-Ray Space Telescope (Fermi) during the first 18 months of the mission. These observations correspond to the recent period of low solar activity when the emission induced by cosmic rays is brightest. For the first time, the high statistical significance of the observations allows clear separation of the two components: the point-like emission from the solar disk due to cosmic ray cascades in the solar atmosphere, and extended emission from the inverse Compton scattering of cosmic ray electrons on solar photons in the heliosphere. The observed integral flux (>100 MeV) from the solar disk is (4.6 +/- 0.2 [statistical error] +1.0/-0.8 [systematic error]) x10^{-7} cm^{-2} s^{-1}, which is ~7 times higher than predicted by the "nominal" model of Seckel et al. (1991). In contrast, the observed integral flux (>100 MeV) of the extended emission from a region of 20 deg radius centered on the Sun, but excluding the disk itself, (6.8 +/-0.7 [stat.] +0.5/-0.4 [syst.]) x10^{-7} cm^{-2} s^{-1}, along with the observed spectrum and the angular profile, are in good agreement with the theoretical predictions for the inverse Compton emission.