Energetic Particle Flux Variations around magnetic storm and huge earthquake

A megathrust earthquake with Mw 9.0 occurred in the North-western Pacific Ocean on March 11, 2011. From the energetic particle flux from WIND, CLUSTER and GOES in different L locations, some variation can be found around the earthquake. Among the three satellites, WIND is used to identify solar activity, and GOES is used to detect the changes from ground source. And during the same period, a magnetic storm with intensity -80nT occur. In order to validate the particle flux variation, multi-parameters relationship is compared. The results show that: (1) all energetic fluxes variation can reflect the solar activity. The far ones are connected with the F10.7 and the near ones are connected with Dst/Kp. (2) The energetic particle fluxes give a scarp change in all energy bands at the beginning coupling period and when the space recovers to be quite, the fluxes will have a long decreasing tail from high to low energy. (3) The coseismic and after effect have been detected in GOES and the pre-seismic emission should exist because the bigger decreasing fluxes in GOES are responding to the period with smaller Kp.

environment is governed principally by the interaction of energetic charged particles with electric and magnetic fields in space. In particular, near the Earth, most of these charged particles derive their energy ultimately from the Sun, directly or through the interaction of the solar wind with the Earth's magnetosphere [Russell, et al, 2016]. Waveparticle interactions also play a fundamental role in energy exchange between plasma waves and radiation belt electrons [Walt, 1994]. Meanwhile, the similar process, but in the opposite direction, has been suggested. Before strong earthquakes, some EM waves, especially the VLF/ELF wave, can be emitted from the lithosphere [Merzer et al., 1997;Molchanov et al., 1998;Zhang X et al., 2014], penetrates into the ionosphere through atmosphere [Rozhnoi et al., 2008 and references therein], and couples with the ionospheric plasma and energetic particles [Nemec, et al., 2009].
Energetic particle variation, called precipitation or enhancement, has been taken as the pre-seismic precursor since 1980s [Galper et al., 1989]. After more and more attentions were paid on the relation between particles and earthquakes in terms of a number of parameters, including time difference, spatial shift according to case study, statistical analysis or some modeling based on wave-particle coupling theory [Voronov et al., 1990;Galper et al., 1992Galper et al., , 1995Pustovetov et al., 1993;Aleksandrin et al., 2003;Sgrigna et al., 2005;Huang et al., 2010;Li et al., 2012;Fidani, 2010;Zhang et al., 2013;Wang et al., 2014].
All these studies showed that before strong earthquakes, energetic particle flux would be affected from 300 to 2000 km altitudes [Wang et al., 2014;Zhang et al., 2013]. However, when a magnetic storm and a strong earthquake happened at the same time, what does the energetic particle response? In this paper, we are devoted to exam whether the pre-seismic flux variation could be found at higher altitudes. So, the higher altitudes data from WIND, Cluster II and GOES are collected, and the time period are set around the occurrence of the M9.0 Tohoku earthquake.

Space weather background and satellite observation 2.1 Space weather background
Several space weather indexes are shown in Figure 1, including f 0.7, Kp, Dst, AU, AL and solar Lyman-alpha. Kp, the geomagnetic three-hourly index, is considered a proxy for the energy input from the solar wind to Earth. During the period, the high value occurred on March 10-12.
Dst index represents the axially symmetric disturbance magnetic field at the dipole equator on the Earth's surface. Major disturbances in Dst are negative, namely decreases in the geomagnetic field. These field decreases are produced mainly by the equatorial current system in the magnetosphere, usually referred to as the ring current. Only on March 11, the Dst decreased lower than -80nT.
The trend of the AU and AL is similar to Dst. And the solar Lymann-alpha is almost same to F10.7.

Jianping Huang et al.
For all the energetic fluxes, three "layers" can be found: the undisturbed region (blue region), the weak increased region (green region) and the core enhancement region (red region). All 4 payloads recorded the changes. The shape at the increased layer is sudden scarp and then weak slope, as shown in SOSP, SFSP and EHSP. The shape in the core enhancement region is like "Gauss shape". The center data of Gauss shape is March 10 for all payloads and also is March 11 in EHSP and ELSP. An indentation, as shown in SOSP, ELSP and not clear in SFSP and EHSP, can also be found in Figure 2. The difference is the starting-ending date and energy range, as listed in Table 1. Table 1 listed the variation date and energy for all payloads. For ELSP, no obvious changes can be seen in the increased layer. The date in SFSP and EHSP is several hours earlier than that in SOSP and lasted longer, too.
And for the enhanced layer, March 10 is the same for all payloads and March 11 is found in EHSP and ELSP.
3 Particle flux around storm and earthquake

CLUSTER II observatory
CLUSTER II is composed of four identical spacecraft flying in a tetrahedral formation, with an aim to study the impact of the Sun's activity on the Earth's space environment. Each orbit took approximately 57 hours to complete. Figure 3 shows the RAPID (Research with Adaptive Particle Imaging Detectors) particle flux from March 1 to March 16, 2011. Figure 3 shows that normally the flux keeps steady and only increase quickly when they pass the perigee which is close to the polar region. The flux increases from March 8 to March 10 and then decreases to normal level.

GOES
The    The electron fluxes from GOES 13 and GOES 15 approximately present the same character (Figure 4). In the lower energy, such as 40eV, 75eV, the periodic daily pump can be found, but in higher energy, it is not so obvious.
As shown in Figure 4, only three decreasingly rapid fluctuations on March 10 and March 11 are found from March 7 to March 17. The first two are before the earthquake and the third one is after the earthquake. For the middle one, the change amplitude is biggest. For the after one, the amplitude is smallest.

Analysis and Discussions
From Figure 1 to 4, each parameter has its variation and it is inconvenient to judge whether the variation is connected with the huge earthquake or not. Figure  Considering the F10.7 and particle flux in WIND, they go in synchronization trend. For GOES, the situation is a little complex. For the 1 st flux drop on March 10, the Kp is 4.3, and the corresponding change in particle flux is from ~6000 to ~20. For the second flux drop on March 11, the kp is 5.3, and the particle flux is from ~1000 to ~2. For the third one, the kp is 5.7 and the flux change is from ~6000 to ~100. In summary, the relative change ratio is 300, 500 and 60 but the spatial Kp is 4.3, 5.3 and 5.7. And based on previous research [He, et al, 2011], the changing amplitude Jianping Huang et al. is consistent with the geomagnetic storm intensity. So generally the change amplitudes of the previous two flux should be smaller than the third one. According to GPS TEC and infrasonic record, the coseismic ionospheric effect was recorded one hour after the earthquake (Hao, et al, 2012), then there is no after-seismic effect on the 3 rd flux drops in GOES because particle flux drop is 19 hours later after the earthquake. So the third drop is only the effect of high Kp. Then there should be some other source which leads to the bigger drop during the T3-T4 and T5-T6.
According to the analysis above, the only source may come from the pre-seismic electromagnetic emission of the coming huge earthquake on March 11. What's more, the duration of the lower flux values could also sustain the conclusion. For the third Kp, it is just a pulse. For the first, the duration is almost 3 hours, and for the second, the duration is almost 4.5 hours. During those durations, some local HF changes could be found which is far different from the spatial index frequency. So the possible pre-seismic effect should be considered for such two decreasing.
In addition, for the T3-T4 and T8-T9, there is no more variation after the kp decreases. While after T5-T6 and T7, the fluxes in GOES occur, which should be the coseismic effect to the energetic particles.

Conclusions
Using the various altitudes observations on WIND, CLUSTER and GOES, the high energetic particle fluxes are compared around the Tohoku huge earthquakes. The results indicate that: All satellite with particle detectors could record the solar activity. And in low orbit satellite, GOES, could also record the pre-and co-seismic particle precipitation.
Particles with the different energy could response to the same source and the responding time and amplitude are different.
Although the spatial background is not quite, the pre-seismic information could be detected. And with more cases, the real pre-seismic energetic particle flux changes could be quantified.