Available theoretical and experimental data (from the huge GLR of 1956, see Fig.1, unfortunately not well measured due to lack of appropriate detectors at this time) on the GLEs confirm proton acceleration up to 20 GeV (Toptigin, 1983; Dorman, 2004). The stochastic acceleration in the flares (Petrosian, 2006) and shock acceleration in corona and interplanetary space (Gang and Zank, 2003) are the two theories aimed to explain the origin and mechanisms of the particle acceleration at the Sun. Middle and high-latitude neutron monitors cannot be used for the reconstruction of the primary energy spectra above 5 GeV due to very weak fluxes and relatively small sizes of the detectors. 

Figure 1. Strongest SEP events of last 60 years

Therefore, recent years surface particle detectors measuring Extensive Air Showers (EAS) were implemented for the investigation of the highest energy solar protons and ions (Ryan, 1999; Ding, 2001; Poirier and D'Andrea, 2002; Chilingarian et al., 2003a). Due to their large surface area and solid angle and high efficiency of the registration of the charged particles, these detectors provide valuable information about the solar proton fluxes well above 5 GeV. 

The Aragats Multidirectional Muon Monitor (AMMM, see Fig.2)a is located at (40.25°N, 44.15°E, cutoff rigidity 7.6 GV) and on altitude 3200 m above sea level; statistical accuracy of 3-min time series of is  ≈0.17%, more sensitive than the neutron monitor 18NM64, located at the same altitude. The AMMM consists of 45 (in 2006 enlarged to 100) plastic scintillators with detecting surface of 1m2 and thickness of 5 cm each. The detector AMMM is located in the underground hall of the ANI experiment (Chilingarian et al., 2003b) under 15 m of soil and concrete, plus 12 cm. of iron bars. Only muons with energies larger than 5 GeV can reach underground hall (muon traversal simulation in 2b), red curve. 5 GeV muons are efficiently produced by primary protons of energy 35-50 GeV if we assume the power-law differential energy spectrum with spectral index of g = -2.7 for Galactic Cosmic Rays, and proton energies of 20-30 GeV if we assume spectral index g  = -4 to -5 (Chilingarian et al., 2005; Zazyan and Chilingarian, 2006). 

During GLE #69 on January 20, 2005 from 7:02 to 7:05 UT, AMMM detects a large flux enhancement, see Fig. 2c. 

Figure 2. a)Aragats Multidirectional Muon Monitor (AMMM); b) simulation of the muon flux traversal; c)detection of flux enhancement during GLE N 69 (1-min time series).

We compare AMMM observation other Aragats Space-Environmental Center (ASEC) monitors (Chilingarian et al., 2006a) and other world-wide monitors, see parameters of the monitors in Table 1, where types, heights, area, cutoff rigidity and geographic coordinate of monitors are presented. Statistical significance (for 3-min time series) is given for peaks occurred at 7:02 UT.

Table 3. Characteristics of the particle detectors registered the GLE #69 at 20 January 2005 

Detectors	Altitude(m)	Surface(m2)	Rigidity GV	Statistical significance	Geographic coordinate
NANM 18NM64	2000	18	7.6	3.7	40.25oN, 44.15oE
ANM 18NM64	3200	18	7.6	1.2	40.25oN, 44.15oE
ASNT-8 channels	3200	4 (60cm thick) 4 (5cm thick)	7.6	0.2 1.5	40.25oN, 44.15oE
AMMM	3200	42	7.6	3.93	40.25oN, 44.15oE
CARPET/Baksan	1700	196	5.7	19	43.28oN, 42.69oE
Tibet YBJ NM 28NM64	4300	28	14.1	12	30.11oN, 90.53oE

Enhancement of the count rate was seen from 7:02 till 7:04 UT with maximum at 7:03 UT. Three out of the 45 one m² scintillators of the AMMM were not operational at the time, therefore only 42 m² of muon detectors were in use to measure the high energy muon flux. The estimated mean count rate of the Galactic Cosmic Rays (GCR) as measured by the 42 m² of the AMMM detector is 123,818 particles per min. The additional signal at 7:03 UT equals to 863 particles or enhancement of 0.70%. Taking into account that the standard deviation of 1 min data is 352 (0.29%) the significance of the one-minute peak at 7:03 UT was 2.5σ  

Figure 3. Detection of the GLE 69 by different Aragats detectors

To emphasize the peak in the AMMM time series we group the 1 min date in 3-min time-intervals (see Fig. 3a). The mean count rate of GCR equals 371,454 particles per 3 min. The additional signal at 7:02 equals 2394 or enhancement of 0.644%. If we adopt the Poisson standard deviation for the 3-min time series 0.164% (see detailed discussion on the determination of the significance of detected enhancement in Chilingarian et al., 2006b) we come to the significance of 3.93 for the 3 min peak at 7:02-7:05 UT. The excess count rate registered at AMMM during the interval 7:02-7:05 UT corresponds to the flux (3.1 ± 0.8) · 10^-5 muons/cm²/s. 

In Figs. 3c and 3d the count rate enhancements measured by the Aragats Neutron Monitor (ANM), located at 3200 m ASL and Nor-Amberd Neutron Monitor (NANM) located at 2000 m ASL are presented (both neutron monitors are 18NM64 type). From the figures we can see that the enhancement at the neutron monitors started ≈3 min earlier than the peak detected by the AMMM and in the interval 6:59-7:45 both ANM and NANM show at least two peaks having significance higher than 3σ. The 5 cm thick plastic scintillators of upper layer of the Aragats Solar Neutron Telescope (ASNT) is sensitive to charged particles with energies greater than 7 MeV. As we can see in Fig. 3b at the same time 6:59-7:45 ASNT also detect several significant peaks. Analogous patterns were detected by the neutron monitors from the world- wide network (Flueckiger et al., 2005). 

The energies of the primary solar protons giving rise to the secondary neutrons (registered by the neutron monitors) and low energy charged particles (registered by surface scintillator detectors) are smaller than the energies of the primary proton that create the 5 GeV muons in the atmosphere. Therefore, we conclude that maximal solar proton energy at 7:12-7:45 was less comparing with 7:02-7:05 when pronounced peak in >5 GeV muon time series was detected. Of course, absence of signal in the AMMM also can be due anisotropic solar protons flux. However, despite the 20 January event was extremely anisotropic at the GLE onset, very soon after onset solar proton flux became rather isotropic (Plainaki et al., 2007; Moraal et al., 2005).

Figure 4. Comparison of the time series of the particle detector sensitive to the highest energies of solar particles: CARPET (energy range >6 GeV), Tibet NM (>13 GeV) and AMMM (>20 GeV). 

The 20 January GLE was detected by several EAS detectors, measuring shower charge particles (mostly muons and electrons) (D'Andrea and Poirier, 2005; Ryan, 2005) and by Tibet YBJ neutron monitor (Miyasaka, 2005); all ensuring registration of highest primary proton energies of 10- 15 GeV. We can see in Fig. 4 rather good agreement of the time series profiles. CARPET and YBJ NM demonstrate high significance peaks in the same time at 7:02. Smaller significance values of AMMM comparing with CARPET and YBJ NM is explained by the much higher threshold of AMMM and large index of the proton flux energy spectrum g= 4-5 (Bieber et al., 2005; Miyasaka, 2005).