Ground level enhancement of the solar cosmic rays on January 20, 2005. - PowerPoint PPT Presentation

About This Presentation
Title:

Ground level enhancement of the solar cosmic rays on January 20, 2005.

Description:

Fig.3. GLE 69 as observed on subpolar NMs McMurdo and Thule. At the first 40 minutes ... at McMurdo significantly exceeds that at Thule even in 10 hrs after the onset. ... – PowerPoint PPT presentation

Number of Views:34
Avg rating:3.0/5.0
Slides: 16
Provided by: ero3
Category:

less

Transcript and Presenter's Notes

Title: Ground level enhancement of the solar cosmic rays on January 20, 2005.


1
Ground level enhancement of the solar cosmic
rays on January 20, 2005.
  • A.V. Belov (a), E.A. Eroshenko (a), H.
    Mavromichalaki (b), C. Plainaki(b), V.G. Yanke
    (a) .
  • Institute of Terrestrial Magnetism, Ionosphere
    and Radio Wave Propagation RAS (IZMIRAN), 142190,
    Troitsk, Moscow region, RUSSIA
  • Nuclear and Particle Physics Section, Physics
    Department, Athens University Pan/polis-Zografos
    15771 Athens, GREECE.
  • rus-eroshenko E-abs2-SH15-poster

2
Abstract
  • Unexpectedly giant proton event toward the end
    of solar cycle was recorded on 20 January 2005 by
    the worldwide network of neutron monitors. The
    GLE was associated with the flare X7.1 on 20
    January in AR720 (N12 W58) started at 636 UT and
    occurred on the background of relatively quiet
    geomagnetic activity. The flux of the first
    relativistic protons reached Earth at 650 UT by
    very narrow beam and had a very hard spectrum.
    The peak of the CR variations reached several
    thousands of percentages at southern polar
    stations.
  • The characteristics of the cosmic ray energy
    spectrum, anisotropy, differential and integral
    fluxes as well, were obtained by the data from
    about 40 neutron monitors using anisotropic and
    compound models of solar cosmic ray variations.

3
One of the greatest GLE
Fig. 1. Relative count rate variations at some
neutron monitors during the greatest GLEs.
4
Large and long anisotropy in GLE69
Fig.3. GLE 69 as observed on subpolar NMs McMurdo
and Thule. At the first 40 minutes the great
anisotropy is evident, but in fact the
enhancement at McMurdo significantly exceeds that
at Thule even in 10 hrs after the onset.
Fig. 2. The visual anisotropy in CR variations at
four NMs CAPS-Cape Shmidt, MCMD-McMurdo, NAIN,
THUL-Thule.
5
Data and Method GLE model
Solar Cosmic Ray Variations
where
are the coupling coefficients
axis-symmetric anisotropy function
Variations of the solar CR flux
Consequently
Contribution of the isotropic part of the SCR
spectrum
Contribution of the anisotropic part of the SCR
spectrum
6
Energy spectrum changes
The first particles came from the Sun by the
narrow beam and had very hard spectrum with an
index -0.65. In some minutes after the onset a
spectrum of the solar CR jumped to become soft
and during the next 5 hours its index changed
only within the -3.0 - 4.0 range. In the first
5-minute interval high energy particles dominated
in the flux whereas just before the 700 UT the
number of low energy particles essentially
enhanced.
Fig.4. Spectra of the solar CR during the first
and second 5-minute intervals after the onset.
7
Mean Flux and Energy Spectrum
Fig. 5. Behavior of the power law energy
spectrum index together with the mean (averaged
by the all directions) particle flux with 1 GeV
energy..
8
Upper Proton Energy
Only in the first minutes the model selects
sufficiently high upper energy, after this the
modeling Eu is small and underestimated. On the
other side, it is a real reflection of the small
contribution of the high energy particles
(gt3GeV) in the observed GLE.
Fig. 6. Behavior of the upper energy Eu,
obtained by model (circles) and observed by
neutron monitors (triangles).
9
Maximal, Minimal and Mean Fluxes
Fig. 7. Behavior of the minimal, maximal and mean
fluxes of solar CR with 1 GeV energy. Anisotropy
decreases quickly at the first time, but it
remains sufficiently noticeable during many hours
after the onset. .
10
Differential Fluxes
Fig. 8. Behavior of differential fluxes of solar
cosmic rays with different energies. Simultaneous
maximum in broad energy range testifies the
arrival of all energies at the same time in the
first minutes.
11
Integral Fluxes
Fig. 9. Integral fluxes of solar CR of
different energies obtained from the NM data
using the GLE model. The fluxes of 300 and 100
MeV are the extrapolation, but there is a good
agreement for maximal flux gt100 MeV with the
satellite observations.
Fig. 10. Behavior of integral fluxes of solar
protons with different energiesby the satellite
observations (GOES measurement).
20 January 2005, UT
12
Anisotropy and source location
Fig.12. Behavior of latitude and longitude of
maximum solar particle flux.
Fig.11. Behavior of the solar CR anisotropy
magnitude and coefficient n, characterizing a
width of angular distribution of anisotropic
flux. .
13
Angular distribution
Fig. 13. Angular distributions at 3 first 5-min
intervals and half an hour later.
14
Longitudinal distribution
Fig. 14. Longitudinal distributions in 3 first
intervals and on the late phase of GLE in
geographic coordinates. At the beginning the beam
is very narrow and then quickly widens. Later
the anisotropy changes by a complicated way. In a
whole the wide angular distribution is
characteristic for the late phase, but model
selects sufficiently often also a narrow beam
even in 8-9 hours after the onset
15
Conclusions
  • The record enhancement in the counting rate at
    some southern polar NMs on 20 January 2005 ranges
    this event among the greatest GLEs. Nevertheless,
    high energy particles (gt3 GeV) turned out to be
    of much less amount than in 1956 and 1989 events.
  • The first particles came by very narrow beam and
    had a very hard spectrum. Already in some minutes
    after the onset the spectrum became soft and kept
    its form during the several hours with index
    -3.0-4.0.
  • The solar CR flux was anisotropic not less than
    11 hours. The changes of anisotropy parameters
    along the time seems to be related to the
    variability of the interplanetary magnetic field.
  • It is possible and necessary to improve the
    obtained results.
Write a Comment
User Comments (0)
About PowerShow.com