The differential energy spectrum of cosmic rays is power-law in nature (on a double logarithmic scale - an inclined straight line) (minimum energies - yellow zone, solar modulation, average energies - blue zone, GCR, maximum energies - purple zone, extragalactic CRs)
Cosmic rays are elementary particles and atomic nuclei moving with high energies in outer space.
Basic information
Cosmic ray physics considered to be part high energy physics And particle physics.
Physics of cosmic rays studies:
- processes leading to the emergence and acceleration of cosmic rays;
- cosmic ray particles, their nature and properties;
- phenomena caused by cosmic ray particles in outer space, and.
Studying the flows of high-energy charged and neutral cosmic particles falling on the boundary of the Earth's atmosphere is the most important experimental task.
Classification according to the origin of cosmic rays:
- in the Galaxy
- in interplanetary space
Primary It is customary to call extragalactic and galactic rays. Secondary It is customary to call particle flows passing and transforming in the Earth’s atmosphere.
Cosmic rays are a component of natural radiation (background radiation) on the Earth's surface and in the atmosphere.
Before the development of accelerator technology, cosmic rays served as the only source of high-energy elementary particles. Thus, the positron and muon were first found in cosmic rays.
The energy spectrum of cosmic rays consists of 43% of the energy of protons, another 23% of the energy of helium (alpha particles) and 34% of the energy transferred by other particles.
By particle number, cosmic rays are 92% protons, 6% helium nuclei, about 1% heavier elements, and about 1% electrons. When studying sources of cosmic rays outside the proton-nuclear component, it is mainly detected by the flux of gamma rays it creates, and the electronic component is detected by the synchrotron radiation it generates, which falls in the radio range (in particular, at meter waves - when emitted in the magnetic field of the interstellar medium ), and with strong magnetic fields in the area of the cosmic ray source - and to higher frequency ranges. Therefore, the electronic component can also be detected by ground-based astronomical instruments.
Traditionally, particles observed in cosmic rays are divided into the following groups: (respectively, protons, alpha particles, light, medium, heavy and superheavy). A feature of the chemical composition of primary cosmic radiation is the anomalously high (several thousand times) content of group L nuclei (lithium, beryllium, boron) compared to the composition of stars and interstellar gas. This phenomenon is explained by the fact that the mechanism of generation of cosmic particles primarily accelerates heavy nuclei, which, when interacting with protons of the interstellar medium, decay into lighter nuclei. This assumption is confirmed by the fact that cosmic rays have a very high degree of isotropy.
History of cosmic ray physics
The first indication of the possibility of the existence of ionizing radiation of extraterrestrial origin was obtained at the beginning of the 20th century in experiments studying the conductivity of gases. The detected spontaneous electric current in the gas could not be explained by ionization arising from the natural radioactivity of the Earth. The observed radiation turned out to be so penetrating that a residual current was still observed in the ionization chambers, shielded by thick layers of lead. In 1911-1912, a number of experiments were carried out with ionization chambers on balloons. Hess discovered that radiation increases with altitude, whereas ionization caused by the radioactivity of the Earth should decrease with altitude. Colherster's experiments proved that this radiation is directed from top to bottom.
In 1921-1925, the American physicist Millikan, studying the absorption of cosmic radiation in the Earth's atmosphere depending on the observation altitude, discovered that in lead this radiation is absorbed in the same way as gamma radiation from nuclei. Millikan was the first to call this radiation cosmic rays. In 1925, Soviet physicists L.A. Tuvim and L.V. Mysovsky measured the absorption of cosmic radiation in water: it turned out that this radiation was absorbed ten times less than the gamma radiation of nuclei. Mysovsky and Tuwim also discovered that the intensity of radiation depends on barometric pressure - they discovered the “barometric effect”. D.V. Skobeltsyn's experiments with a cloud chamber placed in a constant magnetic field made it possible to “see”, due to ionization, traces (tracks) of cosmic particles. D. V. Skobeltsyn discovered showers of cosmic particles. Experiments in cosmic rays made it possible to make a number of fundamental discoveries for the physics of the microworld.
In 1932, Anderson discovered the positron in cosmic rays. In 1937, Anderson and Neddermeyer discovered muons and indicated the type of their decay. In 1947, pi mesons were discovered. In 1955, the presence of K-mesons, as well as heavy neutral particles - hyperons, was established in cosmic rays. The quantum characteristic “strangeness” appeared in experiments with cosmic rays. Experiments in cosmic rays raised the question of parity conservation, discovered processes of multiple generation of particles in nucleon interactions, and made it possible to determine the value of the effective cross section for the interaction of high-energy nucleons. The advent of space rockets and satellites led to new discoveries - the discovery of the Earth (1958, (S.N. Vernov and A.E. Chudakov) and, independently of them in the same year, Van Allen), and allowed the creation of new research methods galactic and intergalactic spaces.
Streams of high-energy charged particles in near-Earth space
There are several types of cosmic rays in near-Earth space (NES). Stationary ones usually include galactic cosmic rays (GCRs), albedo particles and the radiation belt. Non-stationary ones include solar cosmic rays (SCR).
Galactic cosmic rays (GCRs)
Galactic cosmic rays (GCRs) consist of nuclei of various chemical elements with kinetic energy E more than several tens of MeV/nucleon, as well as electrons and positrons with E>10 MeV. These particles come into interplanetary space from the interstellar medium. The most likely sources of cosmic rays are considered to be flares and the resulting ones. The electromagnetic fields of pulsars accelerate charged particles, which are then scattered by interstellar magnetic fields. It is possible, however, that in the area E<100 МэВ/нуклон частицы образуются за счет ускорения в межпланетной среде частиц и . Дифференциальный энергетический спектр ГКЛ носит степенной характер.
Secondary particles in the Earth's magnetosphere: radiation belt, albedo particles
Ultra-high energy cosmic rays
The energy of some particles exceeds the GZK (Greisen - Zatsepin - Kuzmin) limit - the theoretical energy limit for cosmic rays of 5·10 19 eV, caused by their interaction with photons of the cosmic microwave background radiation. Several dozen such particles were recorded by the AGASA observatory per year. These observations do not yet have a sufficiently substantiated scientific explanation.
Detection of cosmic rays
For a long time after the discovery of cosmic rays, the methods for registering them did not differ from the methods for registering particles in accelerators, most often gas-discharge counters or nuclear photographic emulsions raised into the stratosphere or into outer space. But this method does not allow systematic observations of high-energy particles, since they appear quite rarely, and the space in which such a counter can conduct observations is limited by its size.
Modern observatories operate on different principles. When a high-energy particle enters the atmosphere, it interacts with air atoms in the first 100 g/cm² to produce a flurry of particles, mainly pions and muons, which in turn produce other particles, and so on. A cone of particles is formed, which is called a shower. Such particles move at a speed exceeding the speed of light in air, resulting in the Cherenkov glow, which is recorded. This technique makes it possible to monitor areas of the sky covering hundreds of square kilometers.
Boris Arkadyevich Khrenov,
Doctor of Physical and Mathematical Sciences, Research Institute of Nuclear Physics named after. D. V. Skobeltsyn Moscow State University. M. V. Lomonosova
“Science and Life” No. 10, 2008
Almost a hundred years have passed since cosmic rays were discovered - streams of charged particles coming from the depths of the Universe. Since then, many discoveries related to cosmic radiation have been made, but many mysteries still remain. One of them is perhaps the most intriguing: where do particles with an energy of more than 10 20 eV come from, that is, almost a billion trillion electron volts, a million times greater than what will be obtained in the most powerful accelerator - the Large Hadron Collider? What forces and fields accelerate particles to such monstrous energies?
Cosmic rays were discovered in 1912 by the Austrian physicist Victor Hess. He was an employee of the Radium Institute in Vienna and conducted research on ionized gases. By that time, they already knew that all gases (including the atmosphere) are always slightly ionized, which indicated the presence of a radioactive substance (like radium) either in the gas or near a device measuring ionization, most likely in the earth’s crust. Experiments with lifting an ionization detector in a balloon were conceived to test this assumption, since gas ionization should decrease with distance from the earth's surface. The answer was the opposite: Hess discovered some radiation, the intensity of which increased with altitude. This suggested the idea that it came from space, but it was possible to finally prove the extraterrestrial origin of the rays only after numerous experiments (W. Hess was awarded the Nobel Prize only in 1936). Let us remember that the term “radiation” does not mean that these rays are of a purely electromagnetic nature (like sunlight, radio waves or X-rays); it was used to discover a phenomenon whose nature was not yet known. And although it soon became clear that the main component of cosmic rays is accelerated charged particles, protons, the term was retained. The study of the new phenomenon quickly began to produce results that are usually considered to be “the cutting edge of science.”
The discovery of very high-energy cosmic particles immediately (long before the proton accelerator was created) raised the question: what is the mechanism for accelerating charged particles in astrophysical objects? Today we know that the answer turned out to be non-trivial: a natural, “cosmic” accelerator is radically different from man-made accelerators.
It soon became clear that cosmic protons, flying through matter, interact with the nuclei of its atoms, giving birth to previously unknown unstable elementary particles (they were observed primarily in the Earth’s atmosphere). The study of the mechanism of their birth has opened a fruitful path for constructing a taxonomy of elementary particles. In the laboratory, they learned to accelerate protons and electrons and produce huge flows of them, incomparably denser than in cosmic rays. Ultimately, it was experiments on the interaction of particles that received energy in accelerators that led to the creation of a modern picture of the microworld.
In 1938, French physicist Pierre Auger discovered a remarkable phenomenon - showers of secondary cosmic particles that arise as a result of the interaction of primary protons and nuclei of extremely high energies with the nuclei of atmospheric atoms. It turned out that in the spectrum of cosmic rays there are particles with an energy of the order of 10 15 –10 18 eV - millions of times more than the energy of particles accelerated in the laboratory. Academician Dmitry Vladimirovich Skobeltsyn attached particular importance to the study of such particles and immediately after the war, in 1947, together with his closest colleagues G. T. Zatsepin and N. A. Dobrotin, organized comprehensive studies of cascades of secondary particles in the atmosphere, called extensive air showers (EAS) . The history of the first studies of cosmic rays can be found in the books of N. Dobrotin and V. Rossi. Over time, the school of D.V. Skobeltsyna grew into one of the most powerful in the world and for many years determined the main directions in the study of ultra-high-energy cosmic rays. Her methods made it possible to expand the range of energies under study from 10 9 –10 13 eV, recorded on balloons and satellites, to 10 13 –10 20 eV. Two aspects made these studies particularly attractive.
Firstly, it became possible to use high-energy protons created by nature itself to study their interaction with the nuclei of atmospheric atoms and decipher the finest structure of elementary particles.
Secondly, it became possible to find objects in space capable of accelerating particles to extremely high energies.
The first aspect turned out to be not as fruitful as hoped: studying the fine structure of elementary particles required much more data on the interaction of protons than cosmic rays can provide. At the same time, an important contribution to the understanding of the microworld was made by studying the dependence of the most general characteristics of the interaction of protons on their energy. It was during the study of EASs that a feature was discovered in the dependence of the number of secondary particles and their energy distribution on the energy of the primary particle, associated with the quark-gluon structure of elementary particles. These data were later confirmed in experiments at accelerators.
Today, reliable models of the interaction of cosmic rays with the nuclei of atmospheric atoms have been constructed, which have made it possible to study the energy spectrum and composition of their primary particles of the highest energies. It became clear that cosmic rays play no less a role in the dynamics of the Galaxy’s development than its fields and flows of interstellar gas: the specific energy of cosmic rays, gas and magnetic field is approximately equal to 1 eV per cm 3. With such a balance of energy in the interstellar medium, it is natural to assume that the acceleration of cosmic ray particles most likely occurs in the same objects that are responsible for heating and releasing gas, for example, in novae and supernovae during their explosion.
The first mechanism of cosmic ray acceleration was proposed by Enrico Fermi for protons chaotically colliding with magnetized clouds of interstellar plasma, but could not explain all the experimental data. In 1977, Academician Hermogenes Filippovich Krymsky showed that this mechanism should accelerate particles in supernova remnants much more strongly at shock wave fronts, the speeds of which are orders of magnitude higher than the speeds of clouds. Today it has been reliably shown that the mechanism of acceleration of cosmic protons and nuclei by a shock wave in the shells of Supernovae is most effective. But it is unlikely to be able to reproduce it in laboratory conditions: acceleration occurs relatively slowly and requires enormous amounts of energy to retain accelerated particles. In supernova shells, these conditions exist due to the very nature of the explosion. It is remarkable that the acceleration of cosmic rays occurs in a unique astrophysical object, which is responsible for the synthesis of heavy nuclei (heavier than helium) actually present in cosmic rays.
In our Galaxy, there are several known Supernovae less than a thousand years old that have been observed with the naked eye. The most famous are the Crab Nebula in the constellation Taurus (“The Crab” is the remnant of the Supernova explosion in 1054, noted in the eastern chronicles), Cassiopeia-A (observed in 1572 by the astronomer Tycho Brahe) and the Kepler Supernova in the constellation Ophiuchus (1680). The diameters of their shells today are 5–10 light years (1 light year = 10 16 m), that is, they are expanding at a speed of the order of 0.01 the speed of light and are located at distances of approximately ten thousand light years from the Earth. The shells of Supernovae (“nebulae”) were observed in the optical, radio, x-ray and gamma-ray ranges by the Chandra, Hubble and Spitzer space observatories. They reliably showed that acceleration of electrons and protons, accompanied by X-ray radiation, actually occurs in the shells.
About 60 supernova remnants younger than 2000 years could fill interstellar space with cosmic rays with a measured specific energy (~1 eV per cm 3), while less than ten of them are known. This shortage is explained by the fact that in the plane of the Galaxy, where stars and supernovae are concentrated, there is a lot of dust, which does not transmit light to the observer on Earth. Observations in X-ray and gamma rays, for which the dust layer is transparent, have made it possible to expand the list of observed “young” supernova shells. The most recent of these newly discovered shells was Supernova G1.9+0.3, observed with the Chandra X-ray telescope beginning in January 2008. Estimates of the size and expansion rate of its shell indicate that it flared up approximately 140 years ago, but was not visible in the optical range due to the complete absorption of its light by the dust layer of the Galaxy.
The data on Supernovae exploding in our Milky Way Galaxy is supplemented by much richer statistics on Supernovae in other galaxies. Direct confirmation of the presence of accelerated protons and nuclei is gamma radiation with high energy photons resulting from the decay of neutral pions - products of the interaction of protons (and nuclei) with the source matter. Such high-energy photons are observed using telescopes that detect the Vavilov-Cherenkov glow emitted by secondary EAS particles. The most advanced instrument of this type is a six-telescope array created in collaboration with HESS in Namibia. The Crab's gamma rays were the first to be measured, and its intensity became the measure of intensity for other sources.
The obtained result not only confirms the presence of a mechanism for the acceleration of protons and nuclei in a Supernova, but also allows us to estimate the spectrum of accelerated particles: the spectra of “secondary” gamma rays and “primary” protons and nuclei are very close. The magnetic field in the Crab and its size allow the acceleration of protons to energies of the order of 10 15 eV. The spectra of cosmic ray particles in the source and in the interstellar medium are somewhat different, since the probability of particles leaving the source and the lifetime of particles in the Galaxy depend on the energy and charge of the particle. Comparing the energy spectrum and composition of cosmic rays measured near Earth with the spectrum and composition at the source made it possible to understand how long particles travel among stars. There were significantly more lithium, beryllium and boron nuclei in cosmic rays near the Earth than in the source - their additional number appears as a result of the interaction of heavier nuclei with interstellar gas. By measuring this difference, we calculated the amount X the substance through which cosmic rays passed while wandering in the interstellar medium. In nuclear physics, the amount of matter that a particle encounters on its path is measured in g/cm2. This is due to the fact that in order to calculate the reduction in the flux of particles in collisions with nuclei of matter, it is necessary to know the number of collisions of a particle with nuclei that have different areas (sections) transverse to the direction of the particle. By expressing the amount of matter in these units, a single scale of measurement is obtained for all nuclei.
Experimentally found value X~ 5–10 g/cm2 allows you to estimate the lifetime t cosmic rays in the interstellar medium: t ≈ X/ρ c, Where c- particle speed approximately equal to the speed of light, ρ ~10 –24 g/cm 3 - average density of the interstellar medium. Hence the lifetime of cosmic rays is about 10 8 years. This time is much longer than the time of flight of a particle moving at a speed With in a straight line from the source to the Earth (3·10 4 years for the most distant sources on the side of the Galaxy opposite us). This means that the particles do not move in a straight line, but experience scattering. Chaotic magnetic fields of galaxies with induction B ~ 10 –6 gauss (10 –10 tesla) move them around a circle with a radius (gyroradius) R = E/3 × 10 4 B, where R in m, E- particle energy in eV, V - magnetic field induction in gauss. At moderate particle energies E
Approximately in a straight line, only particles with energy will come from the source E> 10 19 eV. Therefore, the direction of particles with energies less than 10 19 eV that create EASs does not indicate their source. In this energy region, all that remains is to observe the secondary radiation generated in the sources themselves by protons and cosmic ray nuclei. In the observable energy region of gamma radiation ( E
The idea of cosmic rays as a “local” galactic phenomenon turned out to be true only for particles of moderate energies E
In 1958, Georgiy Borisovich Christiansen and German Viktorovich Kulikov discovered a sharp change in the appearance of the energy spectrum of cosmic rays at an energy of the order of 3·10 15 eV. At energies below this value, experimental data on the spectrum of particles were usually presented in a “power-law” form so that the number of particles N with a given energy E was considered inversely proportional to the energy of the particle to the power of γ: N(E) = a/Eγ (γ is the differential spectrum indicator). Up to an energy of 3·10 15 eV, the indicator γ = 2.7, but upon transition to higher energies the energy spectrum experiences a “break”: for energies E> 3·10 15 eV γ becomes 3.15. It is natural to associate this change in the spectrum with the approach of the energy of accelerated particles to the maximum possible value calculated for the acceleration mechanism in Supernovae. This explanation of the break in the spectrum is also supported by the nuclear composition of primary particles in the energy range 10 15 –10 17 eV. The most reliable information about it is provided by complex EAS installations - “MGU”, “Tunka”, “Tibet”, “Cascade”. With their help, one obtains not only information about the energy of primary nuclei, but also parameters depending on their atomic numbers - the “width” of the shower, the ratio between the number of electrons and muons, between the number of the most energetic electrons and their total number. All these data indicate that with an increase in the energy of primary particles from the left boundary of the spectrum before its break to the energy after the break, their average mass increases. This change in the mass composition of particles is consistent with the model of particle acceleration in Supernovae - it is limited by the maximum energy, which depends on the charge of the particle. For protons, this maximum energy is of the order of 3·10 15 eV and increases in proportion to the charge of the accelerated particle (nucleus), so that iron nuclei are effectively accelerated up to ~10 17 eV. The intensity of particle flows with energy exceeding the maximum decreases rapidly.
But the registration of particles with even higher energies (~3·10 18 eV) showed that the spectrum of cosmic rays not only does not break, but returns to the form observed before the break!
Measurements of the energy spectrum in the “ultra-high” energy region ( E> 10 18 eV) are very difficult due to the small number of such particles. To observe these rare events, it is necessary to create a network of detectors for the flow of EAS particles and the Vavilov-Cherenkov radiation and ionization radiation (atmospheric fluorescence) generated by them in the atmosphere over an area of hundreds and even thousands of square kilometers. For such large, complex installations, locations are chosen with limited economic activity, but with the ability to ensure reliable operation of a huge number of detectors. Such installations were built first over areas of tens of square kilometers (Yakutsk, Havera Park, Akeno), then hundreds (AGASA, Fly's Eye, HiRes), and finally, installations of thousands of square kilometers are now being created (Pierre Auger Observatory in Argentina, Telescopic installation in Utah, USA).
The next step in the study of ultra-high-energy cosmic rays will be the development of a method for detecting EASs by observing atmospheric fluorescence from space. In cooperation with several countries, Russia is creating the first space EAS detector, the TUS project. Another such detector is expected to be installed on the International Space Station ISS (JEM-EUSO and KLPVE projects).
What do we know today about ultra-high energy cosmic rays? The lower figure shows the energy spectrum of cosmic rays with energies above 10 18 eV, which were obtained using the latest generation installations (HiRes, Pierre Auger Observatory) together with data on cosmic rays of lower energies, which, as shown above, belong to the Milky Way Galaxy. It can be seen that at energies 3·10 18 –3·10 19 eV the differential energy spectrum index decreased to a value of 2.7–2.8, exactly the same as that observed for galactic cosmic rays, when particle energies are much lower than the maximum possible for galactic accelerators . Does this not indicate that at ultra-high energies the main flow of particles is created by accelerators of extragalactic origin with a maximum energy significantly higher than the galactic one? The break in the spectrum of galactic cosmic rays shows that the contribution of extragalactic cosmic rays changes sharply upon transition from the region of moderate energies 10 14 –10 16 eV, where it is approximately 30 times less than the contribution of galactic ones (the spectrum indicated by the dotted line in the figure), to the region of ultra-high energies where it becomes dominant.
In recent decades, numerous astronomical data have been accumulated on extragalactic objects capable of accelerating charged particles to energies much higher than 10 19 eV. An obvious sign that an object of size D can accelerate particles to energy E, is the presence throughout this object of a magnetic field B such that the gyroradius of the particle is less D. Such candidate sources include radio galaxies (emitting strong radio emissions); nuclei of active galaxies containing black holes; colliding galaxies. All of them contain jets of gas (plasma) moving at enormous speeds, approaching the speed of light. Such jets play the role of shock waves necessary for the operation of the accelerator. To estimate their contribution to the observed intensity of cosmic rays, it is necessary to take into account the distribution of sources over distances from the Earth and the energy losses of particles in intergalactic space. Before the discovery of background cosmic radio emission, intergalactic space seemed “empty” and transparent not only to electromagnetic radiation, but also to ultra-high energy particles. The density of gas in intergalactic space, according to astronomical data, is so small (10 –29 g/cm 3) that even at enormous distances of hundreds of billions of light years (10 24 m) particles do not encounter the nuclei of gas atoms. However, when it turned out that the Universe is filled with low-energy photons (approximately 500 photons/cm 3 with energy E f ~10 –3 eV), remaining after the Big Bang, it became clear that protons and nuclei with energy greater E~5·10 19 eV, the Greisen-Zatsepin-Kuzmin (GZK) limit, must interact with photons and lose b O most of your energy. Thus, the overwhelming part of the Universe, located at distances of more than 10 7 light years from us, turned out to be inaccessible for observation in rays with an energy of more than 5·10 19 eV. Recent experimental data on the spectrum of ultra-high energy cosmic rays (HiRes installation, Pierre Auger Observatory) confirm the existence of this energy limit for particles observed from Earth.
As you can see, it is extremely difficult to study the origin of ultra-high energy cosmic rays: the majority of possible sources of cosmic rays of the highest energies (above the GZK limit) are so far away that the particles lose the energy acquired at the source on their way to Earth. And at energies less than the GZK limit, the deflection of particles by the magnetic field of the Galaxy is still large, and the direction of arrival of particles is unlikely to be able to indicate the position of the source on the celestial sphere.
In the search for sources of ultra-high energy cosmic rays, an analysis of the correlation of the experimentally measured direction of arrival of particles with sufficiently high energies is used - such that the fields of the Galaxy slightly deflect the particles from the direction towards the source. Previous generation installations have not yet provided convincing data on the correlation of the direction of arrival of particles with the coordinates of any specially selected class of astrophysical objects. The latest data from the Pierre Auger Observatory can be considered as a hope for obtaining data in the coming years on the role of AGN-type sources in the creation of intense particle flows with energies on the order of the GZK limit.
Interestingly, the AGASA installation received indications of the existence of “empty” directions (those where there are no known sources), along which two or even three particles arrive during the observation. This aroused great interest among physicists involved in cosmology - the science of the origin and development of the Universe, inextricably linked with the physics of elementary particles. It turns out that some models of the structure of the microcosm and the development of the Universe (Big Bang theory) predict the preservation in the modern Universe of supermassive elementary particles with a mass of the order of 10 23 -10 24 eV, of which matter should consist at the earliest stage of the Big Bang. Their distribution in the Universe is not very clear: they can either be uniformly distributed in space, or “attracted” to massive regions of the Universe. Their main feature is that these particles are unstable and can decay into lighter ones, including stable protons, photons and neutrinos, which acquire enormous kinetic energies - more than 10 20 eV. Places where such particles are preserved (topological defects of the Universe) may turn out to be sources of protons, photons or ultra-high energy neutrinos.
As in the case of galactic sources, the existence of extragalactic ultra-high-energy cosmic ray accelerators is confirmed by data from gamma-ray detectors, for example, the HESS telescopes, aimed at the above extragalactic objects - candidates for cosmic ray sources.
Among them, the most promising were active galactic nuclei (AGNs) with gas jets. One of the most well-studied objects at the HESS installation is the M87 galaxy in the constellation Virgo, at a distance of 50 million light years from our Galaxy. At its center there is a black hole, which provides energy to the processes near it and, in particular, to the giant jet of plasma belonging to this galaxy. The acceleration of cosmic rays in M87 is directly confirmed by observations of its gamma radiation, the energy spectrum of photons with an energy of 1–10 TeV (10 12 –10 13 eV), observed at the HESS installation. The observed gamma-ray intensity from M87 is approximately 3% of the intensity of the Crab. Taking into account the difference in distance to these objects (5000 times), this means that the luminosity of M87 exceeds the luminosity of the Crab by 25 million times!
Particle acceleration models generated for this object indicate that the intensity of particles accelerated in M87 could be so great that even at a distance of 50 million light years, the contribution from this source could produce the observed intensity of cosmic rays with energies above 10 19 eV.
But here’s a mystery: in modern data on EASs towards this source there is no excess of particles with an energy of the order of 10 19 eV. But won't this source appear in the results of future space experiments, at such energies when distant sources no longer contribute to the observed events? The situation with a break in the energy spectrum can be repeated again, for example at an energy of 2·10 20 . But this time the source should be visible in measurements of the direction of the primary particle's trajectory, since energies > 2·10 20 eV are so high that the particles should not be deflected in galactic magnetic fields.
As we see, after a century of studying cosmic rays, we are again waiting for new discoveries, this time ultra-high energy cosmic radiation, the nature of which is still unknown, but can play an important role in the structure of the Universe.
Literature:
1) Dobrotin N.A. Cosmic rays. - M.: Publishing house. USSR Academy of Sciences, 1963.
2) Murzin V.S. Introduction to Cosmic Ray Physics. - M.: Publishing house. Moscow State University, 1988.
3) Panasyuk M. I. Strangers of the Universe, or Echoes of the Big Bang. - Fryazino: “Vek2”, 2005.
4) Rossi B. Cosmic rays. - M.: Atomizdat, 1966.
5) Khrenov B.A. Relativistic meteors// Science in Russia, 2001, No. 4.
6) Khrenov B.A. and Panasyuk M.I. Messengers of space: far or near?// Nature, 2006, No. 2.
7) Khrenov B.A. and Klimov P.A. Opening expected// Nature, 2008, No. 4.
Cosmic rays (radiation) are particles that fill interstellar space and constantly bombard the Earth. They were discovered in 1912 by the Austrian physicist Hess using an ionization chamber in a balloon. The maximum energies of cosmic rays are 10 21 eV, i.e. are many orders of magnitude higher than the energies available to modern accelerators (10 12 eV). Therefore, the study of cosmic rays plays an important role not only in cosmic physics, but also in particle physics. A number of elementary particles were first discovered in cosmic rays (positron - Anderson, 1932; muon () - Neddermeyer and Anderson, 1937; pion () - Powell, 1947). Although cosmic rays contain not only charged but also neutral particles (especially many photons and neutrinos), charged particles are usually called cosmic rays.
When discussing cosmic rays, it is necessary to clarify which rays we are talking about. The following types of cosmic rays are distinguished:
1. Galactic cosmic rays - cosmic particles coming to Earth from the depths of our Galaxy. They do not contain particles generated by the Sun.
2. Solar cosmic rays - cosmic particles generated by the Sun.
The flux of galactic cosmic rays bombarding the Earth is approximately isotropic and constant in time and amounts to 1 particle/cm 2 sec (before entering the Earth's atmosphere). The energy density of galactic cosmic rays is 1 eV/cm 3, which is comparable to the total energy of electromagnetic radiation from stars, the thermal motion of interstellar gas and the galactic magnetic field. Thus, cosmic rays are an important component of the Galaxy.
Composition of galactic cosmic rays:
Nuclear component- 93% protons, 6.5% helium nuclei,<1% более тяжелых ядер (т.е. отвечает распространенности ядер во Вселенной).
Electrons. Their number is 1% of the number of cores.
Positrons. Their number is 10% of the number of electrons.
Antihadrons are less than 1%.
The energies of galactic cosmic rays cover a huge range - at least 15 orders of magnitude (10 6 -10 21 eV). Their flux for particles with E>10 9 eV decreases rapidly with increasing energy. The energy spectrum of the nuclear component, excluding low energies, obeys the expression
n(E) = n o E - , (15.5)
ln o is a constant, and 2.7 at E<10 15 ýÂ è 3.1-3.2 ïðè E>10 15 eV. The energy spectrum of the nuclear component is shown in Fig. 15.6.
The flow of ultra-high energy particles is extremely small. Thus, on average, no more than one particle with an energy of 10 20 eV falls on an area of 10 km 2 per year. The nature of the spectrum for electrons with energies >10 9 eV is similar to that shown in Fig. 15.6. The flux of galactic cosmic rays has remained unchanged for at least 1 billion years.
Galactic cosmic rays are obviously of non-thermal origin. Indeed, maximum temperatures (10 9 K) are reached at the center of stars. In this case, the energy of thermal motion of particles is 10 5 eV. At the same time, particles of galactic cosmic rays reaching the Earth's vicinity mainly have energies >10 8 ýÂ.
Rice. 15.6. Energy spectrum of the nuclear component of space
rays. Energy is given in the center of mass system.
There are good reasons to believe that cosmic rays are generated mainly by supernova explosions (other sources of cosmic rays are pulsars, radio galaxies, quasars). In our Galaxy, supernova explosions occur on average at least once every 100 years. It is easy to calculate that to maintain the observed energy density of cosmic rays (1 eV/cm 3), it is enough for them to transfer only a few percent of the explosion power. Protons, heavier nuclei, electrons and positrons ejected during supernova explosions are further accelerated in specific astrophysical processes (they will be discussed below), acquiring energy characteristics inherent in cosmic rays.
There are practically no metagalactic rays in the composition of cosmic rays, i.e. who entered our Galaxy from outside. All observed properties of cosmic rays can be explained based on the fact that they are formed, accumulated and retained in our Galaxy for a long time, slowly flowing into intergalactic space. If cosmic particles moved in a straight line, they would leave the Galaxy several thousand years after their origin. Such a rapid leak would lead to irreparable losses and a sharp decrease in the intensity of cosmic rays.
In fact, the presence of an interstellar magnetic field with a highly entangled configuration of field lines forces charged particles to move along complex trajectories (this movement resembles the diffusion of molecules), increasing the residence time of these particles in the Galaxy by thousands of times. The age of the bulk of cosmic ray particles is estimated at tens of millions of years. Cosmic particles of ultra-high energies are weakly deflected by the galactic magnetic field and leave the Galaxy relatively quickly. This may explain the break in the spectrum of cosmic rays at an energy of 310 15 V.
Let us dwell very briefly on the problem of cosmic ray acceleration. Cosmic ray particles move in the rarefied and electrically neutral cosmic plasma. There are no significant electrostatic fields capable of accelerating charged particles due to the potential difference between different points of the trajectory. But electric fields of inductive and pulsed types can arise in plasma. Thus, an inductive (vortex) electric field appears, as is known, with an increase in the magnetic field strength over time (the so-called betatron effect). Particle acceleration can also be caused by their interaction with the electric field of plasma waves in regions of intense plasma turbulence. There are other acceleration mechanisms that we do not have the opportunity to dwell on in this course. A more detailed examination shows that the proposed acceleration mechanisms are capable of ensuring an increase in the energy of charged particles ejected during supernova explosions from 10 5 to 10 21 V.
Charged particles emitted by the Sun - solar cosmic rays - are a very important component of cosmic radiation bombarding the Earth. These particles are accelerated to high energies in the Sun's upper atmosphere during solar flares. Solar flares are subject to specific time cycles. The most powerful ones repeat with a period of 11 years, the less powerful ones with a period of 27 days. Powerful solar flares can increase the flux of cosmic rays falling on Earth from the Sun by 10 6 times compared to the galactic one.
Compared to galactic cosmic rays, solar cosmic rays contain more protons (up to 98-99% of all nuclei) and, accordingly, fewer helium nuclei (1.5%). They have practically no other nuclei. The content of Z2 nuclei in solar cosmic rays reflects the composition of the solar atmosphere. The energies of solar cosmic ray particles vary in the range of 10 5 -10 11 eV. Their energy spectrum has the form of a power function (15.5), where - decreases from 7 to 2 as the energy decreases.
All the above characteristics of cosmic rays refer to cosmic particles before entering the Earth’s atmosphere, i.e. to the so-called primary cosmic radiation. As a result of interaction with the nuclei of the atmosphere (mainly oxygen and nitrogen), high-energy particles of primary cosmic rays (primarily protons) create a large number of secondary particles - hadrons (pions, protons, neutrons, antinucleons, etc.), leptons (muons, electrons, positrons, neutrinos) and photons. A complex multi-stage cascade process develops. The kinetic energy of secondary particles is spent mainly on ionization of the atmosphere.
The thickness of the earth's atmosphere is about 1000 g/cm2. At the same time, the range of high-energy protons in air is 70-80 g/cm 2 , and that of helium nuclei is 20-30 g/cm 2 . Thus, a high-energy proton can experience up to 15 collisions with atmospheric nuclei and the probability of reaching sea level for the primary proton is extremely small. The first collision usually occurs at an altitude of 20 km.
Leptons and photons appear as a result of weak and electromagnetic decays of secondary hadrons (mainly pions) and the production of e - e + -pairs by -quanta in the Coulomb field of nuclei:
ÿäðî + ÿäðî + e - +e + .
Thus, instead of one primary particle, a large number of secondary ones arise, which are divided into hadronic, muonic and electron-photon components. An avalanche-like increase in the number of particles can lead to the fact that at the maximum of the cascade their number can reach 10 6 -10 9 (at the energy of the primary proton >10 14 eV). Such a cascade covers a large area (many square kilometers) and is called wide atmospheric shower(Fig. 15.7).
After reaching maximum dimensions, the cascade decays mainly due to energy loss due to ionization of the atmosphere. It is mainly relativistic muons that reach the Earth's surface. The electron-photon component is absorbed more strongly and the hadronic component of the cascade almost completely “dies out”. In general, the flux of cosmic ray particles at sea level is approximately 100 times less than the flux of primary cosmic rays, amounting to about 0.01 particles/cm 2 ñåê.
K. l. resemble a highly rarefied relativistic gas, the particles of which practically do not interact with each other, but experience rare collisions with the matter of the interstellar and interplanetary environments and the influence of cosmic. mag. fields. As part of K. l. protons predominate; there are also electrons, nuclei of helium and heavier elements (up to the nuclei of elements with 30). Electrons to K. l. hundreds of times less than protons (in the same energy range). Particles of K. l. have huge kinetics. energies (up to eV). Although the total flux of K. l. near the Earth is small [only 1 particle/(cm 2 s)], their energy density (approx. 1 eV/cm 3) is comparable (within our Galaxy) to the energy density of the total electric magnetic field. radiation from stars, energy of thermal motion of interstellar gas and kinetic. the energy of its turbulent movements, as well as the energy density of the Galaxy’s magnetic field. It follows that K. l. must play a large role in the processes taking place in interstellar space.
Dr. an important feature of K. l. - non-thermal origin of their energy. Indeed, even at a temperature of ~ 10 9 K, apparently close to the maximum for stellar interiors, the average energy of thermal motion of particles is eV. Basic The same number of cosmic ray particles observed near the Earth have energies of 10 8 eV and higher. This means that K. l. acquire energy in specific astrophysical. processes el.-magn. and plasma nature.
Study of K. l. provides valuable information about electromagnetic fields in various areas of outer space. Information “recorded” and “transferred” by cosmic particles. on their way to the Earth, is deciphered during the study - spatiotemporal changes in the flow of cosmic l. under the influence of dynamic el.-magn. and plasma processes in interstellar and near-Earth space.
On the other hand, as a natural source of high-energy particles, K. l. play an irreplaceable role in studying the structure of matter and interactions between elementary particles. Energies of individual particles of cosmic l. so large that they will remain out of competition for a long time in comparison with particles accelerated (to energies of ~ 10 12 eV) by the most powerful laboratory accelerators.
2. Methods for studying cosmic rays
Invading the Earth's atmosphere, primary cosmic rays. destroy the nuclei of the most common elements in the atmosphere - nitrogen and oxygen - and give rise to a cascade process (Fig. 1), in which all currently known elementary particles participate. It is customary to characterize the path traveled by a cosmic particle. in the atmosphere before the collision, the amount of substance in grams contained in a column with a cross section of 1 cm 2, i.e. express the range of particles in g/cm 2 of atmospheric substance. This means that after passing through the atmosphere X(in g/cm2) in a proton beam with initial intensity I 0 the number of protons that did not experience a collision will be equal to , where - avg. particle path. For protons, which make up the majority of primary cosmic rays, in air it is approximately 70 g/cm 2 ; for helium nuclei 25 g/cm 2, for heavier nuclei even less. Protons experience their first collision (70 g/cm2) with atmospheric particles at an average altitude of 20 km. The thickness of the atmosphere at sea level is equivalent to 1030 g/cm2, i.e. corresponds to approximately 15 nuclear ranges for protons. It follows that the probability of reaching the Earth's surface without experiencing collisions is negligible for a primary particle. Therefore, on the surface of the Earth K. l. are detected only by weak ionization effects created by secondary particles.At the beginning of the 20th century. in experiments with electroscopes and ionization. The cameras detected a constant residual ionization of gases caused by some very penetrating radiation. Unlike radiation from environmental radioactive substances, penetrating radiation could not be stopped even by thick layers of lead. The extraterrestrial nature of the detected penetrating radiation was established in 1912-14. Austrian physicist W. Hess, German. scientist W. Kolhurster and other physicists who rose from ionization. balloon cameras. It was found that with increasing distance from the Earth's surface, the ionization caused by cosmic rays increases, for example. at an altitude of 4800 m - four times, at an altitude of 8400 m - 10 times. Extraterrestrial origin of K. l. was finally proven by R. Milliken (USA), who carried out in 1923-26. a series of experiments to study the absorption of K. l. atmosphere (it was he who coined the term “Kl.”).
Nature K. l. up to the 40s. remained unclear. During this time, the nuclear field—the study of the interaction of cosmic rays—developed intensively. with matter, the formation of secondary particles and their absorption in the atmosphere. These studies, carried out using counter telescopes, cloud chambers and nuclear photographic emulsions (raised on sounding balloons into the stratosphere), led, in particular, to the discovery of new elementary particles - the positron (1932), muon (1937), pi -mesons (1947).
Systematic research on the influence of geomagnetic fields on the intensity and direction of arrival of primary cosmic rays. showed that the vast majority of K. l particles. has a positive charge. The east-west asymmetry of cosmic rays is connected with this: due to the deflection of charged particles in the magnetic field. the Earth's field, more particles come from the west than from the east.
The use of photographic emulsions made it possible in 1948 to establish the nuclear composition of primary cosmic rays: traces of nuclei of heavy elements, including iron, were discovered (primary electrons in cosmic rays were first recorded in stratospheric measurements only in 1961). Since the late 40s. The problems of the origin and temporary variations of cosmos gradually came to the fore. (cosmophysical aspect).
Nuclear Phys. research K. l. are carried out mainly using large-area metering installations designed for recording the so-called. extensive atmospheric showers of secondary particles, which are formed during the invasion of one primary particle with energy eV. Basic the purpose of such observations is to study the characteristics of an elementary act of nuclear interaction at high energies. Along with this, they provide information about energy. spectrum of K. l. at eV, which is very important for searching for sources and mechanisms of acceleration of cosmic rays.
Observations by K. l. in cosmophysics aspect are carried out by very diverse methods - depending on the energy of the particles. Variations of K. l. eVs are studied using data from a worldwide network of neutron monitors (the neutron component of cosmic rays), counter telescopes (the muon component of cosmic rays), and other detectors. However, ground-based installations are insensitive to MeV particles due to atmospheric absorption. Therefore, instruments for recording such particles are raised on sounding balloons into the stratosphere to altitudes of 30-35 km.
Extra-atmospheric measurements of cosmic flux. 1-500 MeV are carried out using geophysical. rockets, satellites and other spacecraft. Direct observations of K. l. in interplanetary space have so far been carried out only near the ecliptic plane to a distance of ~ 10 AU. e. from the Sun.
The method of cosmogenic isotopes yielded a number of valuable results. They are formed during the interaction of K. l. with meteorites and space dust, with the surface of the Moon and other planets, with the atmosphere or substance of the Earth. Cosmogenic isotopes carry information about variations in cosmic rays. in the past and about . Based on the content of radiocarbon 14 C in tree rings, it is possible, for example, to study variations in the intensity of cosmic radiation. over the course of several last thousand years. Using other long-lived isotopes (10 Be, 26 Al, 53 Mn, etc.) contained in meteorites, lunar soil, and deep-sea marine sediments, it is possible to reconstruct the picture of changes in the intensity of cosmic rays. for millions of years.
With the development of space technology. technology and radio-chemistry. methods of analysis made it possible to study the characteristics of K. l. along the tracks (traces) created by the nuclei of cosmic rays. in meteorites, lunar matter, in special. target samples exhibited on satellites and returned to Earth, in the helmets of astronauts who worked in outer space, etc. An indirect method of studying K. l is also used. by the ionization effects they cause in the lower part of the ionosphere, especially in polar latitudes. These effects are significant. arr. when solar cosmic rays invade the earth's atmosphere.
3. Cosmic rays near the Earth
Table 1. Relative abundance of nuclei in cosmic rays, in the Sun and stars (on average)
| Element | Solar K.l. | Sun (photosphere) | Stars | Galactic K.l. |
| 1H | 4600* | 1445 | 925 | 685 |
| 2 He (-particle) | 70* | 91 | 150 | 48 |
| 3Li | ? | 0,3 | ||
| 4 Be- 5 B | 0,02 | 0,8 | ||
| 6 C | 0,54* | 0,6 | 0,26 | 1,8 |
| 7N | 0,20 | 0,1 | 0,20 | 0,8 |
| 8 O** | 1,0* | 1,0 | 1,0 | 1,0 |
| 9F | 10 -3 | 0,1 | ||
| 10 Ne | 0,16* | 0,054 | 0,36 | 0,30 |
| 11 Na | ? | 0,002 | 0,002 | 0,19 |
| 12 Mg | 0,18* | 0,05 | 0,040 | 0,32 |
| 13Al | ? | 0,002 | 0,004 | 0,06 |
| 14 Si | 0,13* | 0,065 | 0,045 | 0,12 |
| 15 P- 21 Sc | 0,06 | 0,032 | 0,024 | 0,13 |
| 16 S- 20 Ca | 0,04* | 0,028 | 0,02 | 0,11 |
| 22 Ti- 28 Ni | 0,02 | 0,006 | 0,033 | 0,28 |
| 26 Fe | 0,15* | 0,05 | 0,06 | 0,14 |
* Observational data for the interval = 1-20 MeV/nucleon, the remaining numbers in this column relate mainly to >40 MeV/nucleon. The accuracy of most values in the table as a whole is from 10 to 50%. ** The abundance of oxygen nuclei is taken as unity.
The most important characteristics of K. l. yavl. their composition (distribution of masses and charges), energy. spectrum (distribution by energy) and degree of anisotropy (distribution by direction of arrival). Relative content of nuclei in cosmic l. is given in Table 1. From the table 1 it is clear that in the composition of K. l. galactic origin of much more light nuclei ( Z= 3-5) than in solar K. l. and on average in the stars of the Galaxy. In addition, they contain significantly more heavy poisons (20) compared to their natural abundance. Both of these differences are very important for clarifying the question of the origin of K. l.
Relative numbers of particles with different masses in cosmic liters. are given in table. 2.
Table 2. Composition and some characteristics of cosmic rays with energies of 2.5 GeV/nucleon
It can be seen that protons predominate in the flow of primary cosmic particles, accounting for more than 90% of all particles. In relation to protons, particles make up 7%, electrons ~ 1% and heavy nuclei - less than 1%. These figures refer to particles with an energy of 2.5 GeV/nucleon as measured near the Earth at minimum solar activity, when the observed energies. the spectrum can be considered close to the unmodulated spectrum of cosmic rays. in interstellar space.
Integral energy spectrum of K. l. align="absmiddle" width="145" height="22"> [particles/(cm 2 s)] reflects the dependence of the number of particles I with energy higher ( I 0 is a normalizing constant, +1 is a spectrum indicator, the minus sign indicates that the spectrum has a decreasing character, i.e. with increasing intensity of K. l. decreases). Often they also use a differential representation of the spectrum [particles/(cm 2 s MeV)], which reflects the dependence on the number of particles per unit energy interval (1 MeV).
The differential spectrum, compared to the integral spectrum, allows us to identify more subtle energy details. distribution of K. l. This can be seen from Fig. 2, which shows the differential spectrum of cosmic rays observed near the Earth in the range from approximately 10 6 to eV. Particles of K. l. with energies falling in this interval are influenced by solar activity, therefore the study of energy. spectrum K. l. in the range 10 6 -10 11 eV is extremely important for understanding the penetration of cosmic rays. from interstellar to interplanetary space, interactions of cosmic rays. with interplanetary magnet. field (IMF) and for the interpretation of solar-terrestrial connections.Before the start of extra-atmospheric and extra-magnetospheric observations of cosmic rays. the question of the shape of the differential spectrum in the eV region seemed quite clear: the spectrum near the Earth has a maximum near 400 MeV/nucleon; the unmodulated spectrum in interstellar space must have a power-law shape; There should be no galactic ones in interplanetary space. K. l. low energies. Direct measurements of K. l. in the range from 10 6 to 10 8 eV showed, contrary to expectations, that, starting from approximately = 30 MeV (and below), the intensity of cosmic rays. grows again, i.e. a characteristic dip in the spectrum was discovered. Probably, the failure is the result of increased modulation of K. l. in the eV region, where particle scattering on IMF inhomogeneities is most effective.
It has been established that at eV the spectrum of K. l. is no longer subject to modulation, and its slope corresponds to a value of 2.7 up to eV. At this point the spectrum undergoes a break (the indicator increases to = 3.2-3.3). There are indications that at the same time in the composition of K. l. the proportion of heavy nuclei increases. However, data on the composition of K. l. in this energy region are still very scarce. At align="absmiddle" width="118" height="17"> eV, the spectrum should end abruptly due to the escape of particles into the intergalactic space. space and interactions with photons. The flow of particles in the ultra-high energy region is very small: on average, no more than one eV particle falls on an area of 10 km 2 per year.For K. l. eV is characterized by high isotropy: with an accuracy of 0.1%, the intensity of particles in all directions is the same. At higher energies, anisotropy increases and in the eV range reaches several. tens of % (Fig. 3). Anisotropy ~0.1% with a maximum near 19:00 sidereal time corresponds to the predominant direction of motion of cosmic rays. along magnetic field lines. galactic fields spiral arm in which the Sun is located. With increasing particle energy, the time of maximum shifts to 13 hours of sidereal time, which corresponds to the presence of a cosmic ray drift flow. with eV from the Galaxy across magnetic field lines.
4. Origin of cosmic rays
Due to the high isotropy of cosmic l. Observations near the Earth do not allow us to establish where they are formed and how they are distributed in the Universe. These questions were answered by radio astronomy in connection with the discovery of space exploration. in the radio frequency range Hz. This radiation is created by very high energy electrons as they move through the magnet. Galaxy field.
The frequency at which the intensity of radio emission is maximum is related to the magnetic field strength. fields N and electron energy by the ratio (Hz), where is the pitch angle of the electron (the angle between the electron velocity vector and the vector N). Magn. field of the Galaxy, measured several times. methods, has a value of E. On average, at E and =0.5, eV, i.e. radio-emitting electrons must have the same energies as the main ones. mass of cosmic rays observed near the Earth. These electrons, which are one of the components of cosmic rays, occupy an extended region covering the entire galaxy and is called galactic. halo. In interstellar magnetic In fields, electrons move like other charged particles of high energy - protons and heavier nuclei. The only difference is that, due to their low mass, electrons, unlike heavier particles, intensely emit radio waves and thereby detect themselves in distant parts of the Galaxy, being an indicator of cosmic rays. at all.
In addition to the general galactic Discrete sources of synchrotron radio emission were discovered: shells, the galactic core, . It is natural to expect that all these source objects of cosmic rays.
Until the beginning of the 70s. 20th century many researchers believed that K. l. with align="absmiddle" width="89" height="17"> eV are mainly metagalactic. origin. At the same time, the absence of known galaxies was indicated. sources of particles with up to 10 21 eV and the difficulties associated with the problem of their containment in the Galaxy. In connection with the discovery of pulsars (1967), a number of possible mechanisms for the acceleration of even very heavy nuclei to ultra-high energies were considered. On the other hand, the data obtained indicate that the electrons observed near the Earth are formed and accumulated in the Galaxy. There is no reason to think that protons and heavier nuclei behave differently in this regard. Thus, the galactic theory is justified. origin of K. l.Indirect confirmation of this theory was obtained from data on the distribution of cosmic sources across the celestial sphere. gamma radiation. This radiation arises due to the decay of mesons, which are formed during collisions of cosmic rays. with particles of interstellar gas, as well as due to bremsstrahlung radiation from relativistic electrons during their collisions with particles of interstellar gas. Gamma rays are not affected by magnetism. fields, so the direction of their arrival directly points to the source. In contrast to the almost isotropic distribution of cosmic rays observed inside the Solar System, the distribution of gamma radiation across the sky turned out to be very uneven and similar to the distribution of supernovae across the galaxy. longitude (Fig. 4). The good agreement between the experimental data and the expected distribution of gamma radiation over the celestial sphere serves as strong evidence that the main The source of cosmic rays is supernovae.
Theory of the origin of K. l. relies not only on the hypothesis of galactic the nature of the sources of K. l., but also on the idea that K. l. are retained in the Galaxy for a long time, slowly flowing into the intergalactic. space. Moving in a straight line, K. l. would have left the Galaxy several times later. thousand years after the moment of generation. On the scale of the Galaxy, this time is so short that it would be impossible to compensate for losses with such a rapid leak. However, in the interstellar magnetic field. field with highly entangled lines of force movement of cosmic l. has a complex nature, reminiscent of the diffusion of molecules in a gas. As a result, the leakage time of K. l. from the Galaxy turns out to be thousands of times greater than during rectilinear motion. The above concerns the basic parts of particles K. l. (with eV). Particles with higher energy, the number of which is very small, are weakly deflected by the galactic. mag. field and leave the Galaxy relatively quickly. This is apparently associated with a break in the spectrum of radiation. at eV.
The most reliable estimate of the leakage time of CO l. from the Galaxy is obtained from data on their composition. In K. l. light nuclei (Li, Be, B) are present in very large quantities (compared to the average abundance of elements). They are formed from the heavier nuclei of cosmic rays. when the latter collide with the nuclei of atoms of interstellar gas (mainly hydrogen). In order for light nuclei to be present in the observed quantity, K. l. During their movement in the Galaxy they must pass through a thickness of interstellar matter of approx. 3 g/cm. According to data on the distribution of interstellar gas and remnants of supernova explosions, the age of cosmic rays. does not exceed 30 million years.
In favor of supernovae as the main source of cosmic rays, in addition to radio, x-ray and gamma-ray astronomy data, also indicate estimates of their energy release during flares. Supernova explosions are accompanied by the release of huge masses of gas, forming a large, brightly glowing and expanding shell (nebula) around the exploding star. The full energy of the explosion is spent on radiation and kinetic energy. the energy of gas expansion can reach 10 51 -10 52 erg. In our Galaxy, according to the latest data, supernovae erupt on average at least once every 100 years. If we assign the flare energy of 10 51 erg to this time period, then cf. The flash power will be approx. erg/s. On the other hand, to maintain modern energy density K.l. OK. 1 eV/cm power of K. l sources. at Wed. life time of K. l. in the Galaxy, years should be at least 10 40 erg/s. It follows that in order to maintain the energy density of cosmic l. in modern level is enough for them to receive only a few. % supernova explosion power. However, radio astronomy can only directly detect radio-emitting electrons. Therefore, it cannot yet be definitively stated (although this seems quite natural, especially in the light of the achievements of gamma-ray astronomy) that during supernova explosions a sufficient number of protons and heavier nuclei are also generated. In this regard, the search for other possible sources of K. l. has not lost importance. Of great interest in this regard are pulsars (where, apparently, particle acceleration to ultra-high energies is possible) and the galactic region. nuclei (where explosive processes of much greater power than supernova explosions are possible). However, the generation power of cosmic rays is galactic core does not apparently exceed the total power of their generation during supernova explosions. In addition, most of the cosmic rays formed in the core will leave the galactic disk before reaching the vicinity of the Sun. Thus, we can assume that supernova explosions are phenomena. the main, although not the only source of K. l.
5. Mechanisms of cosmic ray acceleration
The question of possible mechanisms for accelerating particles to energies of ~ 10 21 eV in detail is still far from being finalized. solutions. However, in general terms the nature of the acceleration process is already clear. In an ordinary (non-ionized) gas, the redistribution of energy between particles occurs due to their collisions with each other. In rarefied cosmic In plasma, collisions between charged particles play a very small role, and the change in energy (acceleration or deceleration) of an individual particle is due to its interaction with the electric magnet. fields arising from the movement of all plasma particles surrounding it.
Under normal conditions, the number of particles with an energy noticeably exceeding av. the energy of thermal motion of plasma particles is negligibly small. Therefore, the acceleration of particles should begin practically from thermal energies. In space plasma (electrically neutral) cannot exist any significant electrostatic. fields, which could accelerate charged particles due to the potential difference between points of the field. However, electricity can occur in the plasma. fields of a pulsed or inductive nature. Pulse electric fields appear, for example, when a neutral current layer breaks in the area of magnetic contact. fields of opposite polarity (see). Induction electric the field appears as the magnetic intensity increases. fields over time (betatron effect). In addition to pulsed fields, the initial stage of acceleration can be caused by the interaction of accelerated particles with the electric fields of plasma waves in areas with intense turbulent plasma motion.
In space, apparently, there is a hierarchy of accelerating mechanisms, which work in different combinations or in different sequences depending on the specific conditions in the field of acceleration. Acceleration by pulsed electric field or plasma turbulence contributes to subsequent acceleration by the induction (betatron) mechanism or the Fermi mechanism.
Certain features of the process of particle acceleration in space are associated with the behavior of plasma in magnetic fields. field. Cosmic mag. fields exist in large volumes of space. Particle with charge Ze and impulse p moves in magnetic field H along a curved path with an instantaneous radius of curvature
,
Where R = cp/Ze- mag. particle stiffness (measured in volts), - particle pitch angle. If the field changes little at distances comparable to the value , then the particle trajectory has the form of a helical line winding around the magnetic field line. fields. In this case, the field lines are, as it were, attached to the plasma (frozen into the plasma) - the displacement of any part of the plasma causes a corresponding displacement and deformation of the magnetic field lines. fields, and vice versa. If sufficiently intense motions are excited in the plasma (this situation arises, for example, as a result of a supernova explosion), then there are many such randomly moving sections of the plasma. For clarity, it is convenient to consider them as separate plasma clouds moving relative to each other at high speeds. Basic the mass of plasma particles is held in the clouds and moves with them. However, a small number of high-energy particles, for which the radius of curvature of the trajectory is in mag. The plasma field is comparable to the size of the cloud or exceeds it; when it enters the cloud, it does not remain in it. These particles are only deflected magnetically. field of the cloud, it is as if a particle collides with the cloud as a whole and particles are scattered on it (Fig. 5). Under such conditions, the particle effectively exchanges energy with the entire cloud at once. But kinetic. the energy of the cloud is very high and, in principle, the energy of the accelerated particles can grow unlimitedly until the particle leaves the region with intense plasma movements. This is the essence of statistics. acceleration mechanism proposed by E. Fermi in 1949. Particles are accelerated similarly when they interact with powerful shock waves (for example, in interplanetary space), in particular when two shock waves approach each other, forming reflective magnets. "mirrors" (or "walls") for accelerated particles.
When accelerated by plasma waves, particles with only several energies can be accelerated. times more thermal. The number of such particles is not too small, but the acceleration conditions will significantly depend on the type of particles, which should lead to a strong change in their composition compared to the composition of the initial plasma. The spectrum of accelerated protons, however, in this case can also be ~ exp -(R/R 0).
The betatron mechanism, which is based on the preservation of adiabatic. invariant of particle motion = const, gives a power-law spectrum and is not selective with respect to the type of particles, but its effectiveness is proportional to the magnetic field. particle rigidity ( dR/dt ~ R), i.e. For its action, preliminary acceleration (injection) is necessary.
The Fermi acceleration mechanism gives power-law energy. spectrum, however, it is selective with respect to the type of particles. Acceleration by shock waves in space. plasma also leads to power-law energy. spectrum, and theoretically. calculations give an index of =2.5, which corresponds quite well to the observed shape of the spectrum of cosmic rays. Thus, the acceleration theory, unfortunately, allows for an ambiguous approach to the interpretation of the observed spectra of accelerated particles (in particular, solar cosmic rays).
Acceleration processes by pulsed electric fields near the magnetic zero lines. fields are observed during solar flares, when for several. min particles appear, accelerated to an energy of several. GeV. Near pulsars, in the shells of supernovae in the Galaxy, as well as in extragalactic ones. objects - radio galaxies and quasars - this process can also play a major role. acceleration mechanism or at least the role of the injector. In the latter case, the injected particles are accelerated to max. observed in K. l. energies as a result of interactions with waves and magnetic inhomogeneities. fields in turbulent plasma.
Observations on various scales (Galaxy, Sun, Earth's magnetosphere, etc.) show that particle acceleration occurs in space. plasma wherever there are sufficiently intense inhomogeneous movements and magnetic fields. fields. However, in large numbers and to very high energies, particles can be accelerated only where a very large kinetic force is imparted to the plasma. energy. This is exactly what happens in such grandiose cosmic environments. processes such as supernova explosions, the activity of radio galaxies and quasars.
Along with the huge role of K. l. in astrophysics processes, it is necessary to note their importance for studying the distant past of the Earth (climate changes, evolution of the biosphere, etc.) and for solving some practical problems. modern tasks (ensuring the radiation safety of cosmonauts, assessing the possible contribution of cosmic radiation to meteorological effects, etc.).
Lit.:
Ginzburg V.L., Syrovatsky S.I., Origin of cosmic rays, M., 1963; Miroshnichenko L.I., Cosmic rays in interplanetary space, M., 1973; Dorman L.I., Experimental and theoretical foundations of cosmic ray astrophysics, M., 1975; Toptygin I, N., Cosmic rays in interplanetary magnetic fields, M., 1983.
(L.I. Miroshnichenko)
Cosmic rays (CR) are usually understood as streams of charged relativistic particles, starting from protons and helium nuclei and ending with nuclei of heavier elements up to uranium, generated and accelerated to high and extremely high (up to 10 20 eV) energies outside the Earth. In this case, the flux of particles with energies up to 10 9 eV is dominated by the contribution of the Sun, and particles of higher energies are of galactic (and, possibly, at the highest energies, extragalactic) origin.
Naturally, protons and nuclei do not exhaust the entire variety of radiation coming to Earth from outer space. However, issues related to the study of other components that make up cosmic radiation: electrons, positrons, antiprotons, neutrinos, gamma quanta, as well as various electromagnetic radiation, are not covered here.
The composition of galactic cosmic rays (GCRs) is dominated by protons, with the remaining nuclei accounting for less than 10%. Protons remain the dominant component, at least up to energies of ~1 TeV, although the proportion of nuclei increases with particle energy. Figure 1 compares the relative abundance of nuclei in the CR with the abundance of elements in the solar system (Simpson, 1997). In general, similarity is observed, with two exceptions: the group Li, Be, B and elements from Cl to Mn.
Rice. 1 Representation of elements. Dark dots are cosmic rays, light dots are the Solar System.
As can be seen from the figure, the content of light nuclei in the GCR (with charge Z from 3 to 5) is several orders of magnitude higher than their content in stars. In addition, GCRs are characterized by a significantly greater presence of heavy nuclei (Z>20) compared to their natural abundance. The anomalously high representation of these elements is associated with an additional contribution from the splitting of heavier elements in the interstellar medium. Both of these factors are very significant for clarifying the question of the origin of GCR.
The Sun is also a source of CRs, and fluxes of solar cosmic rays (SCRs), especially during solar flares, can reach very high values, however, the characteristic value of their energy, as a rule, does not exceed 109 eV, while GCRs are distributed over a very wide range of energies from 109 to 1020 eV. Therefore, the division of cosmic rays into galactic and solar cosmic rays reflects the essence of the matter, since both the characteristics and sources of cosmic rays and cosmic rays are completely different. At energies below 10 GeV/nucleon, the GCR intensity measured near the Earth depends on the level of solar activity (more precisely, on the magnetic field changing during solar cycles). 
In the region of higher energies, the GCR intensity is constant in time. According to existing concepts, GCRs themselves end in the energy region between 10 17 and 10 18 eV. Therefore, at energies above 10 18 eV it is more correct to use the designation simply CR, since the origin of cosmic rays of extremely high energies is most likely not connected with the Galaxy. The observed differential CR energy spectrum (Cronin, 1999) is shown in Fig. 2. The spectrum is described by a power law over a very wide energy range from 10 11 to 10 20 eV with a slight change in slope of about 3 10 15 eV (kink, sometimes called knee) and about 10 19 eV (ankle). The integrated CR flux above the ankle is approximately 1 particle per km 2 per year.
Fig.2 Energy spectrum of cosmic rays.
The power-law nature of the CR energy spectrum indicates the non-thermal origin of their energy, and this, in turn, imposes certain requirements on CR sources, which must ensure the formation of a power-law energy spectrum. The maximum energy of CR particles that has been recorded from observations of extensive air showers is 3.10 20 eV, and there are more than 10 events whose energy is >10 20 eV. Such energies can hardly be provided by sources located in our Galaxy. At the same time, the interaction of cosmic rays of extremely high energies with cosmic microwave background radiation with a temperature of 2.75ºK limits the range of distances from which particles with such energies could come to the region of the local supercluster of galaxies, and in it, as in our Galaxy, there are also no objects that can provide acceleration to such high energies. This problem attracts close attention of researchers, and to solve it, installations with huge sensitive areas are created, since the intensity of extremely high-energy particles is extremely low (see Fig. 2).
The energy density carried by cosmic rays is ~1 eV/cm3; the largest contribution to this value, due to the steeply falling spectrum, is made by particles of relatively low energies. Meanwhile, it is significant that the value of the GCR energy density turns out to be comparable with the energy density of the thermal motion of interstellar gas and its turbulent motions, with the density of the total electromagnetic radiation of the stars of our Galaxy and with the energy density contained in the magnetic field of the Galaxy. This means that the role of GCR in the energy balance of processes occurring in the Universe is quite large, and this circumstance should be taken into account by the theory of the origin of cosmic rays (Astrophysics CR, 1990).
The GCR flow is characterized by a high degree of isotropy. The values of the anisotropy coefficient up to 10 14 eV do not exceed 0.1%; with a further increase in energy, the CR anisotropy coefficient increases and reaches several tens of percent at energies >10 19 eV; however, the statistical significance of the experimental results is in the region of ultra-high and extremely high energies (10 15 –10 20 eV), as a rule, is small.
The theory of the origin of GCRs, which could be called completely complete, is currently missing, especially if we bear in mind the origin of GCRs of ultra-high energies (>10 15 eV), although over the past 10–15 years there has been an understanding of the general nature of the processes, in which cosmic rays appear and accelerate, and significant progress has been made. A complete theory of the origin of GCRs should explain the main characteristics of GCRs: the power-law shape of the energy spectrum, the value of the energy density, the mass (chemical) composition of primary CRs, including data on the fluxes of antiprotons, electrons, positrons, gamma rays, the practical constancy of the GCR intensity over time and the very weak their anisotropy. Back in the late 1950s, energy considerations (Ginzburg and Syrovatsky 1963) led to the conclusion that the source of GCRs (at least the bulk of their mass) should be considered supernova explosions in our Galaxy. The quantitative theory of converting the energy of a supernova explosion into the energy spectrum of cosmic rays by accelerating charged particles by shock waves in expanding supernova shells began to develop in the late 1970s (Krymsky, 1977) and has now become generally accepted, although it has not yet received final experimental confirmation. This theory makes it possible to describe the formation of a power-law GCR spectrum up to energies of ~10 15 .Z eV, where Z is the charge of the accelerated ion, and even up to ~10 17 .Z eV (Ptuskin and Zirakashvili, 2005) taking into account the large magnetohydrodynamic turbulence arising from for the instabilities of the CR flux at the early stage of supernova evolution, but additional efforts are needed to understand how particles are accelerated up to energies of 10 20 eV.
The energy spectrum of GCRs and their mass composition, observed near the Earth, are formed as a result of transformation during the passage from sources distributed mainly within the central part of the galactic disk to the solar system located on the periphery of the Galaxy. Since both regular and random magnetic fields exist in the Galaxy, the characteristic strength of which is ~3.10 -6 G, GCR particles propagate along very intricate trajectories, and their movement can be described to a good approximation as diffusion. The main arguments in favor of the presence of diffusion are associated with the almost complete isotropy of the GCR flux and the presence of light nuclei (Li, Be, B) in the GCR flux in quantities hundreds of thousands of times greater than their abundance in the Galaxy. The lifetime of GCRs, i.e., the time they remain in the Galaxy, is ~3.10 7 years, which is 4 orders of magnitude greater than the time required to cross the Galaxy when moving in a straight line. During this time, the range of nuclei of medium elements (C, N, O) will be 5–10 g/cm 2 in the interstellar gas, which is sufficient for the formation of light nuclei. The lifetime of GCRs and the amount of matter they pass through decrease with increasing particle energy; Particles of extremely high energies practically no longer experience diffusion.
The energy spectrum and mass composition of GCRs can be measured either directly, i.e., as a result of direct registration of GCR particles in experiments carried out on balloons and satellites, or using indirect methods based on studying the characteristics of extensive air showers (EAS) occurring as a result of the development of a cascade process in the atmosphere. The advantage of the EAS method is that some shower components can be detected at very large distances from the trajectory of the primary particle that generated the EAS (up to tens of kilometers when recording fluorescence created by charged shower particles in the atmosphere), which achieves a huge increase in the effective detection area of the event . This makes it possible to overcome the inevitable limitations of statistics inherent in direct experiments and which does not allow them to be used to study GCRs above a certain energy threshold, which depends on the geometric factor of the detector. Currently, the record value of energy achieved in experiments on the Proton series satellites (1968) is ~ 2.10 15 eV. For most direct experiments, this threshold is still significantly lower, so that the boundary between direct and indirect experiments lies between energies of 10 14 –10 15 eV. However, the price for using the advantages of indirect methods is the need to determine the energy and mass number of the primary particle based on the results of the development of the cascade in the atmosphere, which is associated with significant uncertainty even if it is known exactly how the elementary act of interaction occurs. Meanwhile, our information regarding hadron-nucleon interactions is limited to an energy of 2.10 15 eV (the equivalent energy of the Tevatron in a laboratory system). At the same time, it should be emphasized that the same uncertainty would be inherent in experiments carried out using ionization calorimeters on satellites and balloons if these experiments were aimed at an energy region for which there is no experimental data on hadron-nucleon interactions.
2. METHODS FOR STUDYING COSMIC RAYS
Due to the large extent of the CR spectrum in terms of energy and its steeply falling nature, it is necessary to use various measurement methods.
2.1 Direct methods
Experimental study of GCRs using direct methods suggests the possibility of directly measuring the charge and energy of primary particles. As already mentioned in the Introduction, the upper limit of the energy range within which direct methods can currently be used is approximately 10 15 eV. This limit is determined based on the natural requirement to achieve the minimum acceptable statistical accuracy within a reasonable time of the experiment. Although this value is much smaller than the upper limit of the CL spectrum (~ 10 20 eV), however, in this case, the energy range in which studies are carried out by direct methods extends to 5 orders of magnitude, which leads to the need to use various methods for measuring charge and energy ( or momentum) of the primary particles.
As is known, the Earth's magnetic field can serve as an analyzer of the magnetic rigidity of particles, which in the past made it possible to obtain the first information regarding the energy spectrum of GCRs in the region up to approximately 10 GeV. The range from 10 GeV to 10 15 eV was studied using photographic emulsions, ionization calorimeters, magnetic spectrometers, X-ray emulsion chambers and some other instruments installed on satellites or lifted on cylinders.
An ionization calorimeter is a fairly thick block of material layered with ionization detectors, which allows, using the readings of the detectors, to determine the total ionization created by the cascade generated by the primary particle, and then find the primary energy using either modeling of the cascade process or calibration of the ionization calorimeter at an accelerator. Ideally, the ionization calorimeter should completely absorb the entire cascade created by the primary particle in the substance. However, when placing an ionization calorimeter on a satellite or balloon, such a requirement is not feasible, so the calorimeter can directly measure only part of the energy of the primary particle, and therefore errors in energy measurements increase with increasing particle energy. An ionization calorimeter can exist in a photoemulsion version, and can also be a combination of layers of X-ray emulsion film, used as an ionization detector, measured by the optical density of the blackening of the film, with layers of an absorber; It is also possible to use semiconductor ionization detectors. If the thickness of the calorimeter is small, so that there are only 1-2 layers of ionization detectors, then the calorimeter turns into a so-called push installation (a push is a burst of ionization in the detector during the passage of an avalanche of charged particles). Unlike calorimeters, pusher installations allow one to measure only the number of charged particles at the maximum of the cascade, and not the total ionization created by the cascade.
To measure the charge of the primary particle, as a rule, special detectors are used. These detectors take advantage of the fact that both ionization losses and losses due to Cherenkov radiation are proportional to Z 2 - the square of the charge of the primary particle. This allows for separation by Z either by the magnitude of the ionization losses of the particle, or by the flux of Cherenkov radiation created by the particle (Cherenkov counter).
Research in outer space was started in the 1960s by Grigorov and his colleagues in experiments on the Proton series satellites (Bugakov et al., 1970). In these experiments, the charge and direction of particle motion were determined using Cherenkov counters with plexiglass radiators, and an ionization calorimeter was used to determine the energy (Fig. 3), containing 140 g/cm 2 Pb and 855 g/cm 2 Fe as an absorber between 16 layers ionization chambers (to this day this calorimeter remains a record in weight and luminosity).
Rice. 3 Schematic diagram of the IK-15 spectrometer for studying high-energy cosmic ray particles; M – replaceable graphite and polyethylene targets, ChS – Cherenkov counters, TM – thin graphite targets, DN – detectors of charge and particle direction, IR – ionization chambers, PS – proportional counters.
In experiments on the Proton series satellites, the energy spectrum of all particles at energies of 10 11 –10 15 eV and separately the spectra of protons and α-particles were measured.
The continued development of technology in subsequent years led to the implementation of three large experiments in space: HEAO-3, SOKOL and CRN, in which spectra were measured up to energies of ~1 TeV/nucleon for elements up to iron. Balloon experiments began in the 1970s to measure the spectra of various elements at energies above 100 GeV/nucleon.
Due to the development of the emulsion chamber method, long flights providing greater exposure became possible. A series of experiments were performed: MUBEE, JACEE, RUNJOB. A typical emulsion chamber used for direct measurements of cosmic rays and their interactions at the top of the atmosphere by the JACEE collaboration (Asakimori, 1998) is shown in Fig. 4. 
> This camera was designed to measure primary composition when exposed to above 99.5% atmosphere. The top of the chamber consists of layers of sensitive emulsion separated by layers of plastic. The charge of the incident primary nucleus is measured before its interaction by the degree of darkening of the track in the emulsion. The middle part of the camera is designed to track tracks with minimal chance of interaction. This allows the tracks to diverge sufficiently so that cascades generated by interactions in the calorimetric part of the chamber can be individually measured.
Rice. 4 – Emulsion chamber in the JACEE experiment.
The essential elements of the calorimeter are x-ray films and lead plates. Electromagnetic cascades, generated either directly by electrons or photons, or photons from the decay of π 0 -mesons, develop rapidly in lead, and their energy can be determined by summing the blackening measurements in the X-ray film layers along each cascade. Characteristics of a number of space and balloon experiments, as well as data on future planned experiments, are summarized in Table 1 (Wefel, 2003).
Table 1 Experiments to study the spectra and chemical composition of galactic cosmic rays
| Experiment, Years | Core | Methodology | Energy range, eV | Geom. factor,/m 2 .sr. | Exposure factor /m 2 avg.day |
| Spacecraft | |||||
| Proton 1-4 1965-1968 | All cores H, He | calorimeter | 10 11 - 10 15 | 0.05 - 10 | 5 - 2000 |
| HEAO-3 1979-1980 | 16≤Z≤28 | ionization/Cherenkov | 3.10 10 - 10 13 | 1.2 | 370 |
| HEAO-3 1979-1980 | 4≤Z≤28 | Cherenkovsky detectors | 3.10 10 - 2.10 12 | 0.14 | 33 |
| CRN Spacelab2 1985 | 5≤Z≤26 | Transition radiation detectors | 7.10 11 - 3.10 13 | 0.1- 0.5 0.5 -0.9 | 0.3 -3 |
| FALCON (Space) 1984-1986 | 1≤Z≤26 | calorimeter | 2.10 12 - 10 14 | 0.026 | 0.4 |
| Balloons | |||||
| Ryan et al 1969-1970 | 1≤Z≤26 | calorimeter | 5.10 10 - 2.10 12 | 0.036 | 0.01 |
| JACEE | 1≤Z≤26 | emulsion chamber | 10 12 - 5.10 14 | 2-5 | 107(H,He) 65(Z>2) |
| MUBEE 1975-1987 | 1≤Z≤26 | emulsion camera | 10 13 - 3.10 14 | 0.6 | 22 |
| RUNJOB 1995-1999 | 1≤Z≤26 | emulsion camera |
10 13 - 5.10 14 | 1.6 | 43 |
| ATIC Antarctic 2000-2001 | 1≤Z≤28 | calorimeter | 10 10 - 10 14 | 0.23 | 3.5 |
| ATIC Antarctic 2002-2003 | 1≤Z≤28 | calorimeter | 10 10 - 10 14 | 0.23 | 6.9 |
| TRACER 2004-2005 | 1≤Z≤28 | detectors transition radiation | 10 11 - 3.10 14 | 5 | 70 |
| CREAM 2004-2005 | 1≤Z≤28 | transient detectors radiation/calorimeter | 10 12 - 5.10 14 | 1.4 -0.35 | 35 - 140 |
| New experiments | |||||
| Spacecraft | |||||
| ACCESS | detectors transition radiation | 10 13 - 5.10 15 | 7 - 12 | 7000 - 12000 | |
| (CSTRD) | calorimeter | 10 12 - 10 15 | 0,9 | 900 | |
| PROTON-S | calorimeter | 10 12 - 3.10 16 | 18 | 18000 | |
| INCA | neutron calorimeter | 10 14 - 10 16 | 48 | 48000 | |
| AMS | superconducting | 10 10 - 10 13 | 50 | 50000 | |
In Fig. Figure 5 shows a schematic diagram of the AMS experiment instrument (Casaus et al, 2003).
Rice. 5 Schematic diagram of the AMS device.
When considering the results of measuring the spectra and composition of GCRs using direct methods (see later in the text), statistical limitations of the data are obvious, so qualitative and quantitative improvement of the experimental situation is necessary. Taking into account the falling nature of the GCR energy spectrum, which leads to a sharp drop in the intensity of the GCR flux with an increase in the energy of detected particles, a detector with an area of 1 m 2 at the boundary of the atmosphere will register about 100 events per year with an energy > 10 15 eV. This leads to the conclusion that an energy of ≈ 10 15 eV separates the energy region in which direct methods can be used from the ultra-high energy region, where currently only indirect methods can be used.
2.2 Indirect methods
The ability to obtain information about ultra-high-energy GCRs is due to the existence of the Earth's atmosphere, in which a primary particle develops a hadronic-electromagnetic cascade consisting of a large number of secondary particles and called an extensive air shower (EAS). This name is due to the fact that secondary particles arising as a result of interactions and decays can be detected at sufficiently large distances from the EAS axis - a straight line coinciding with the direction of motion of the primary particle. Depending on the primary energy, EAS detection can occur at distances of the order of hundreds or even thousands of meters from the axis, so that the effective area can reach tens of square kilometers. All this makes it possible to study EASs using a system of isolated detectors placed so as to cover the largest possible area (Christiansen et al. 1975).
To implement the EAS method, large-area detectors are required, designed for long exposures, which is due to the small flux of particles with such energies. The most common way is to build installations on the surface of the earth that can cover areas measured in square kilometers and operate for years.
The EAS method still remains the most powerful method for obtaining information about PCRs with energies above 10 15 eV. It is this method that, up to the highest observed energies of ~ 3.10 20 eV, has provided the majority of data on the main characteristics of PCR: energy spectrum, mass composition and anisotropy (Kalmykov and Khristiansen, 1995).
Historically, the first method used to study EASs was the method of detecting EASs by recording streams of charged particles, and due to its relative simplicity, it is still widespread today. The properties of EASs and methodological issues are described in detail in the review by Greisen (1958), which has not lost its significance to this day. 
The basis of EASs is the hadron cascade in the atmosphere, developing from a primary particle - a proton or nucleus (Fig. 6), interacting at the boundary of the atmosphere.
Rice. 6– EAS development diagram (Haungs, 2003).
As the cascade develops, other EAS components are formed - the electron-photon component, the muon component, as well as optical radiation resulting from the passage of charged particles through the atmosphere (Cherenkov and fluorescent). The most numerous of the charged EAS particles are electrons, which usually include positrons. The number of muons is approximately 10% of the number of electrons (with the number of electrons Ne ≈10 5 –10 6). The number of gamma rays is approximately twice the number of electrons, and hadrons make up ~1% of the total number of particles in EASs.
The development of a shower in the atmosphere occurs in such a way that the number of particles in an EAS first increases, then reaches a maximum, and then decreases as the energy of an increasing number of particles falls below the threshold for further particle formation. EAS particles form a thin disk of relativistic particles. The high-energy hadrons that make up the EAS trunk feed the electromagnetic part of the shower, mainly with photons from the decay of neutral pions. Nucleons and other high-energy hadrons contribute to the hadron cascade. Charged pions and kaons of lower energy decay, contributing to the muon component. (The relationship between decay and interaction depends on energy and depth in the atmosphere.)
With each hadronic interaction, slightly more than a third of the energy is transferred to the electromagnetic component. Since most hadrons interact repeatedly, most of the primary energy gradually turns into the electromagnetic component. Bremsstrahlung radiation of photons by electrons and positrons, as well as the generation of electron-positron pairs by photons lead to the rapid multiplication of particles in electromagnetic cascades, so that the number of electrons and positrons in the shower increases. After the shower passes the maximum, the number of electrons and positrons begins to decrease, since due to the fragmentation of energy between particles, their characteristic energy becomes below critical (Ec ~ 80 MeV), after which the electrons and positrons quickly lose the remaining energy to ionization. Therefore, most of the shower energy is finally dissipated due to ionization losses of electrons and positrons. Except for a small fraction F(E0) energy carried away by muons and neutrinos, primary energy E0 is determined by the total length of the trajectories of all electrons in the atmosphere (track length integral):
Where N(x) is the number of charged particles in the shower at depth x (measured along the shower axis) and α is the energy loss per unit path length in the atmosphere. 
An example of a setup for studying EASs is shown in Fig. 7.
Along with the detection of EAS by the flux of charged particles, methods for detecting EAS have also become widespread, based on the registration of optical radiation accompanying EAS - Cherenkov radiation and ionization glow or fluorescence.
Rice. 7- Installation of KASCADE (Klages et al, 1997).
It is important that the fluxes of both Cherenkov light and fluorescence are determined mainly by the characteristics of electron-photon cascades, which can be calculated with better accuracy than the characteristics of hadron cascades, and therefore the fluxes of Cherenkov radiation and fluorescence are less subject to model dependence. This is an important advantage, although the implementation of the method requires the installation to operate only on clear moonless nights, which reduces the actual experiment time to 5-10% of the astronomical time. Fluorescence detectors are an essential part of the Pierre Auger installation and, with a primary energy of ~ 10 20 eV, make it possible to detect the passage of EAS at a distance of up to 40 km from the detector. Projects are being developed to record the fluorescence created by EAS in the atmosphere using space-based installations.
Interesting data, essential for determining the mass composition of GCRs, is provided by studying the hadronic component of EASs. However, the fluxes of hadrons are significantly inferior to the fluxes of the electron and muon components, and the equipment necessary for detecting hadrons is quite complex (ionization calorimeter) and expensive, so the hadron component is rarely studied in modern installations for detecting EASs.
It seems promising to use X-ray emulsion chambers of large area up to ~1000 m 2 as part of EAS installations (Fig. 8), as in the Pamir experiment (Baiburina et al., 1984), to measure the high-energy central part of EAS, making it possible to register TeV particles with a spatial resolution of 300 µm. 
Rice. 8 Scheme of using an X-ray emulsion chamber (Kempa, 1997).
To obtain information about primary cosmic rays from EAS data, an integrated approach is needed to ensure that the largest possible number of characteristics are found in each shower. Simultaneous registration of the muon component along with the electron component makes it possible to extract information about the mass composition of the primary radiation. For the same purpose, one can use information on the longitudinal development of the electron-photon cascade in the atmosphere, as well as on the functions of the spatial distribution of certain components.
The use of EAS to determine the energy spectrum and mass composition of GCRs is inevitably associated with the need to reconstruct the parameters of the primary particle (energy, mass number, and direction of its arrival) from the responses of the detectors included in the installation. Such a reconstruction is impossible unless one has a model of this phenomenon, based on extrapolation of accelerator data on the characteristics of hadronic interactions to the ultra-high energy region, where such data are absent. Formally, accelerator data now end at an equivalent laboratory energy of 1.8.10 15 eV, but a number of important characteristics of hadron-nucleon interactions and, in particular, hadron-nucleus interactions are known only up to energies of ~1 TeV. Since the models of hadron interactions currently used are phenomenological, then, strictly speaking, the reliability of their predictions cannot be guaranteed outside the energy region within which the model parameters were determined. This circumstance should always be kept in mind when interpreting experimental data obtained by studying EASs.
3. COSMIC RAYS NEAR THE EARTH
3.1 Modulation effects area
The lowest energy particles cannot be observed directly near the Earth because the solar wind prevents these particles from entering our Heliosphere. This heliospheric modulation decreases with increasing energy and leads to a solar cycle of CR intensity variation at low energies. Noticeable changes occur in the intensity and spectrum of GCRs entering the Heliosphere. These changes are primarily associated with the interaction of the cosmic ray flux with the solar wind and magnetic fields frozen into this wind. As a result, the energy spectrum of galactic cosmic rays measured near the Earth differs markedly from the GCR spectrum in the interstellar medium. Figure 9 shows the results of measurements of the spectrum of galactic cosmic rays during time periods corresponding to different phases of solar activity (Heber, 2001). 
Rice. 9 Energy spectrum of various elements measured near the Earth in the year of minimum solar activity (upper curves) and in the year of maximum (lower).
It can be seen that at energies above 10 GeV/nucleon, the GCR intensities in different phases of solar activity differ slightly. At the same time, at energies of ~10 MeV, the intensities of the spectra can differ by an order of magnitude.
When considering various phenomena in the heliosphere over several decades, their determining factor is the 11-year and 22-year cyclicity of the solar process, characterized by a number of clearly established patterns regarding the level of solar activity, the location of active regions on the photosphere, as well as the magnetic field of active formations. The boundary of the modulation region is located at distances of ~100 AU.
Figure 10 shows the modulation of CR intensity in the 11-year solar cycle (Bazilevskaya et al., 2005). The GCR intensity changes in antiphase with the number of sunspots. However, the processes of solar modulation turn out to be quite complex and cannot be reduced only to anti-correlation with the number of sunspots.
The theoretical basis for GCR transport in the heliosphere is the Parker transport equation (Parker, 1965):
Where is the cosmic ray distribution function, R is the hardness, r and t are the distance from the Sun and time, respectively. V – solar wind speed. The right side of the equation contains terms describing particle convection, longitudinal and transverse drift, diffusion, adiabatic energy changes and particle source, respectively. The source of the particles can be any heliospheric source. K is a tensor, the symmetric part of which describes diffusion, and the antisymmetric part of the tensor describes the drift of particles in the heliospheric magnetic field with an average speed V D . In recent years, taking into account diffusion in the direction perpendicular to the magnetic field has become especially important.
Equation (1) is usually solved numerically. Its solution, in principle, makes it possible to obtain modulation values inside the heliosphere. However, the variety of natural processes and connections in which CR are involved is so great that when solving this equation, a problem arises - the need for detailed knowledge of the spatial, temporal and energy dependencies of the main parameters of the equation on the size and geometry of the modulation region.
Rice. 10 Intensity of cosmic rays with energy > 100 MeV at the boundary of the atmosphere in the Murmansk region according to stratospheric measurements. The solid line indicates the CR intensity, the dotted line indicates the number of sunspots.
Due to the complexity of the problem, modulation models based on three-dimensional, energy-dependent numerical simulations have recently been very actively improved. The calculation results can be compared with experimental data obtained on balloons and spacecraft. In (Bonino et al, 2001), using an approximate solution of the transport equation, a differential energy spectrum of protons is presented, depending on the solar modulation parameter M:
Here T is the kinetic energy per nucleon, and E0 is the rest energy of the nucleon. In the same work, experimental data from observations of the spectrum of galactic cosmic rays on balloons and spacecraft were analyzed. 29 different experiments were considered. By comparing the results of calculations using formula (2) with these data, the solar modulation parameters M were determined that best describe the experimental intensity values. (Fig.11) 
Rice. 11 Differential spectra of cosmic rays obtained on the basis of equation (2) for various values of solar modulation M = 390, 600, 820, 1080 MeV (curves 1,2,3,4, respectively) in comparison with experimental data obtained on balloons and space devices during 1965, 1968, 1980 and 1989. respectively.
There is a semi-empirical dynamic model (Nymmik, 2005) that allows one to describe fluxes of GCR particles with Z from 1 to 92 and with energies from 5 to 10 5 MeV/nucleon. The model takes into account the dependence of fluxes on the level of solar activity, as well as the magnitude and direction of the solar magnetic field.
3.2 Energy region 10 11 –10 17 eV
3.2.1 Direct experiments
Above energies of ~10.Z GeV, the modulation caused by the magnetic field of the heliosphere is negligible and it can be considered, to a first approximation, that the spectra of individual elements included in the GCR follow a power law. The same remark is true for all GCR particles. The spectrum index changes at an energy of 3-4 PeV from approximately –2.7 to –3.1, and this break in the spectrum is often called a “knee”. The origin of the knee, discovered almost 50 years ago (Kulikov and Christiansen, 1958), is still a matter of debate. Various possibilities for the occurrence of a kink due to either a change in the nature of GCR propagation in our Galaxy, or a change in the process of particle acceleration, are considered further in Sections 4 and 5. However, it must be emphasized that in both cases the energy at which a kink should occur for nuclei with a charge Z turns out to be proportional to Z.
In Fig. 12, 13, 14 show the results of direct experiments on the study of fluxes of protons, helium nuclei and iron nuclei (Horandel, 2003), as well as approximations constructed according to the table from the same work.
 |
 |
 |
Fig. 12-14 Spectra of protons, helium and iron nuclei
3.2.2 Methodology for determining the energy spectrum and mass composition of GCRs from EAS data
When using EASs as a tool for studying ultrahigh-energy cosmic rays, determining the primary energy and mass composition turn out to be, generally speaking, interrelated. Indeed, the methods used are based either on the simultaneous measurement of several components of an individual EAS at a given level of observation, or on information about its longitudinal development. The development of EAS depends both on the energy of the primary particle that generated the shower and on its mass number. The most widely used method for obtaining information about the mass number of a primary particle is to study the relationship between the number of electrons Ne and the number of muons Nμ. On average, EASs from primary nuclei develop faster in the atmosphere and have a larger number of muons.
The spatial distributions of various EAS components and, in particular, Cherenkov radiation, carry information about the shape of the cascade curve and, therefore, about how quickly the shower develops in the atmosphere. The study of the distributions of arrival times of various EAS components at the observation level (Cherenkov or fluorescent light, muons) also provides information about the actual development of EASs and is used in experimental practice.
Drawing physical conclusions from the analysis of experimentally observed EASs is a rather complex process due to the fact that there are fluctuations associated with the random nature of cascade processes, as well as various kinds of systematic uncertainties that arise during the detection of EASs. In the general case, the characteristics of the primary particle that interest us must be determined with the most accurate consideration of both the fluctuations inherent in cascade processes and all the necessary details of the measurement process.
For the purpose of modeling the EAS development process, a number of Monte Carlo programs have been developed: CORSIKA (Heck et al, 1998), MOCCA (Hillas, 1981), AIRES (Sciutto, 1999) and new ones continue to be developed. Since the direct use of the Monte Carlo method from the energy of the primary particle to the threshold energy of directly detected particles requires significant computer time, at primary energies >10 16 eV, schemes with the introduction of statistical weights are usually used (Hillas, 1997), which can lead to artificial fluctuations. The use of numerical methods makes it possible to significantly reduce the time for calculating the average characteristics of the process, but it turns out to be a much less convenient tool if it is necessary to take into account fluctuations and simulate the process of detecting EAS. Therefore, the most promising direction for the development of computational methods seems to be the synthesis of Monte Carlo approaches and numerical methods (Kalmykov et al, 1997).
To determine the GCR energy spectrum in the region of the first break (10 15 –10 17 eV), it is necessary to have an estimate of the EAS energy, and the best solution to the problem would be a calorimetric type estimate, if possible independent of the mass number of the particle that generated the given shower. Unfortunately, this is not always possible, so different installations use different methods for converting from observed spectra to energy spectra.
Estimating the energy and mass number of a primary particle based on the results of recording fluxes of secondary EAS components reduces to solving the inverse problem. The methods used are divided into two significantly different classes: the use of a deconvolution procedure (unfolding), in which the energy spectrum and mass composition are extracted from experimentally measured spectra for Ne, Nμ, etc., and the use of various methods of pattern recognition theory, where, by comparison with Theoretical distributions assign individual detected EASs to one or another mass number.
The deconvolution method is used to solve the Fredholm integral equation of the 1st kind, which, in relation to the problem at hand, can be written as follows:
Where F(Ne(μ)) is the spectrum of electrons (or muons) experimentally measured by the installation, Ii(E) is the energy spectrum of primary particles belonging to group i (protons, helium nuclei, nuclei of the CNO group, etc. up to nuclei iron), is the probability that a primary particle with energy E and mass number corresponding to group of nuclei i will create a shower with the required number of electrons or muons.
To increase the accuracy of solving the problem, it is desirable to consider simultaneously as many data as possible; for example, when analyzing KASCADE data, spectra of electrons and muons were used in several ranges of zenith angles (Roth et al, 2003). To estimate the energy in the KASCADE experiment, the so-called “truncated” number of muons is used, equal to the integral of the muon density in the range from 40 to 200 m from the EAS axis. As is known, special additional measures are required to obtain a unique solution to the Fredholm integral equation of the 1st kind (regularization (Blobel, 1985), positivity of the transfer function (Gold, 1964) or the requirement of smoothness of the solution (D’Agostini, 1995)). It should also be noted that calculating probability requires large computational costs, and so far the statistics of the theoretical event bank are inferior to the experimental ones. Overcoming this situation requires the development of combined calculation methods.
Pattern recognition can be considered as the task of estimating the density of distributions in a multidimensional space, followed by dividing the area under study into areas, the entry into which is interpreted as the assignment of the primary particle that generated a given EAS to one or another group of nuclei. Theoretically, the best is the so-called Bayesian classifier, which minimizes the probability of classification error (Fukunaga, 1972). However, other methods are also used, in particular the neural network method (Bishop, 1995). The use of individual event classification (Glasmacher et al, 1999) works best when the sample under study a priori contains only two different types of particles (for example, partitioning into light and heavy nuclei). With a larger number of groups, the effectiveness of the method decreases due to an increase in classification error.
3.2.3 GCR energy spectrum according to EAS data
Since the nature of the break in the GCR energy spectrum at an energy of ~ 3·10 15 eV is not yet fully understood, it is currently difficult to propose a computational model that would describe the spectra of individual nuclei, including the break region, and would not raise doubts. The spectra of individual groups of nuclei obtained in the KASCADE experiment (Horandel, 2003) demonstrate the presence of breaks, and the energy of the break turns out to be proportional to the charge of the nucleus. However, the intensities of the individual spectra depend on the interaction model adopted, which cannot currently be definitively established. Nevertheless, analysis of data from direct experiments and installations for studying EASs made it possible to propose a phenomenological kink model (Horandel, 2003), which successfully describes the available experimental data.
The energy dependence of the flow of particles with charge Z is taken in the following form:
Below the break energy EZ, the spectra have the usual power-law form, with γZ depending on Z. This dependence is determined from direct measurement data. At energies much higher than EZ, the spectrum is determined by the exponent γc, with |γc|>|γZ|. The value of εc determines how abruptly the transition from one mode to another occurs. The parameters EZ, γc and εc are determined from the analysis of data from the KASCADE installation.
The most interesting result of this analysis seems to be the following. Despite the presence of a model dependence of the I 0Z values, the spectrum of all particles practically does not reveal such a dependence. Moreover, extrapolation of direct measurement data in accordance with the assumed form of the I Z (E) energy spectra fits well with the results obtained by analyzing data from a large number of EAS installations, especially if some renormalization of the GCR energy spectra reconstructed from EAS data is carried out (see Fig. Fig. 15). In this case, as a rule, a change in energy of only a few percent is sufficient. The optimal values of EZ, γc and εc are equal to: EZ=Z Ep, where Er=(4.51±0.52) PeV; γc=–4.68±0.23; εc=1.87±0.18.
Rice. 15 Differential energy spectra of all particles.
Thus, the indicators of the partial spectra after the break increase by almost 2.0. The value εc≈2 corresponds to the region of transition from γZ to γc, which occupies approximately half an order of magnitude. Taking into account the presence in the GCR of elements up to uranium, which experiences a break at an energy of ~4.10 17 eV, the proposed phenomenological model makes it possible to describe the energy spectrum of the GCR, approximately up to the indicated energy. At high energies it must be assumed that cosmic rays have a different, most likely extragalactic, origin.
3.3 Results of the study of CR anisotropy
One of the main characteristics of CLs is their possible anisotropy. Anisotropy measurements are important from the point of view of identifying the spatial distribution of sources in the Galaxy and the nature of the motion of relativistic charged particles. Information about anisotropy is of particular interest for interpreting the break in the energy spectrum of GCRs at E 0 ≈ 3·10 15 eV.
One of the sources of anisotropy is the anisotropy associated with the peculiar motion of the solar system relative to the total mass of stars, interstellar gas and the large-scale magnetic field of the Galaxy (the Compton-Gätting effect). The resulting anisotropy is of the order of σ ≈3·10 -4. Other reasons for the appearance of anisotropy are due to the general outflow of cosmic rays generated in our Galaxy into metagalactic space without a significant role of the reverse flow and the contribution of individual nearby sources (pulsars, supernova remnants).
Reliable information about the anisotropy of cosmic rays in the Galaxy using ground-based measurements can be obtained only for particles with energies greater than 5·10 11 –10 12 eV, since the movement of particles of lower energies is strongly distorted by the magnetic field of the solar system.
The study of CR anisotropy is usually based on an analysis of the dependence of their intensity I(t) on sidereal time t. The intensity can be represented as a Fourier series:
where A 0 is the isotropic component, ω = 2π/T, T is the duration of the sidereal day, An is the amplitude, and φn is the phase of the nth harmonic. Usually they are limited to finding A1 and φ1, dividing the entire measurement period into separate intervals, during which temperature and barometric differences are relatively small.
(The barometric coefficient is 1% per 1 mm Hg, and the temperature coefficient is about 1% per 10 C. Therefore, when studying anisotropy violation with an error of the order of a percent, accurate accounting of the barometric and temperature effects is necessary.)
From the definition of anisotropy
And the expressions for I(t), neglecting harmonics of the second and higher order, we obtain
The use of diffusion models to calculate anisotropy is limited, since anisotropy can largely be determined by the local structure of the magnetic field near the solar system.
Relationship between the anisotropy value δ and the CR concentration gradient
Arising in the model of isotropic diffusion, it is violated due to the tensor nature of diffusion associated with the “magnetization” of the relativistic CR gas.
The results of anisotropy measurements: the amplitude of the first harmonic A and its phase φ, that is, the direction to the maximum intensity, are shown in Figure 16 (Ambrosio et al, 2003). 
Fig. 16 – CL anisotropy. Amplitude of the first harmonic (a) and its phase (b)
Only the most reliable data are presented, for which A/σ≥3, where σ is the root mean square error. As can be seen from the figure, the amplitude and phase of anisotropy do not show a noticeable dependence on energy up to energy E0≤10 15 eV.
At high energies, the currently available data on CR anisotropy are very uncertain, mainly due to a lack of statistics, and allow one to estimate only the upper limit of the anisotropy. However, apparently, we can talk about a tendency towards an increase in anisotropy and a change in its direction.
At energies E ≥ 10 15 eV, the anisotropy is mainly due to the outflow of GCRs from the Galaxy due to diffusion, and the diffusion coefficient depends on energy as D~E 0 0.6. At these energies, there may be a significant contribution to the anisotropy due to the drift of particles in the regular magnetic field of the Galaxy. Due to the effect of drift (Hall diffusion) of GCRs (Zirakashvili et al. 1991), in the general regular magnetic field of the Galaxy the anisotropy is δ~D(E) and an anisotropy of ~10 -2 is permissible at E0≈10 17 eV.
3.4 Cosmic rays at energies above 10 17 eV
Isolating cosmic rays with energies above 10 17 eV into a separate point is advisable for two reasons. Firstly, the energy of 10 17 eV is the boundary energy of confinement of particles of such energy in the Galaxy by magnetic inhomogeneities, which have a characteristic scale of ~100 pc. Secondly, from an experimental point of view, at these energies there is a transition from compact EAS installations, which make it possible to determine the total number of particles in a shower at the observation level, reflecting the energy of the primary particle, to expanded installations, in which one or another classification parameter is used to find the primary energy .
Most of the data at energies above 10 17 eV were obtained at the EAS installations: Havera Park, Yakutsk, AGASA and using detectors that record fluorescent light from nitrogen atoms excited in the atmosphere: Fly’s Eye and HiRes. Unfortunately, Havera Park, AGASA and Fly’s Eye installations have ceased operation. 
Fig. 17 Differential energy spectrum of CLs with energies above 10 17 eV.
Figure 17 shows the differential energy spectra of PCR at energies above 10 17 eV, measured in Yakutsk (Glushkov et al, 2003), in the AGASA (Sakaki et al, 2001) and HiRes experiments (Abbasi et al, 2005).
It can be seen from the figure that the CL intensity according to the Yakut group data is noticeably higher (2.5 times compared to HiRes), and the spectrum is somewhat steeper.
Based on the entire set of experimental data, the energy spectrum is characterized by the following features: the spectrum steepens to E-3..3 above 10 17.7 eV (dip), and then settles down to E -2.7 at 10 18.5 eV (ancle). The most common interpretation of the ankle is that above 10 18.5 eV a new population of cosmic rays of extragalactic origin begins to dominate the galactic component (Cocconi 1996).
This hypothesis is supported by anisotropy data. At an energy of about 10 17 eV, deviations from isotropy are small. According to data from Havera Park (Lloyd-Evans and Watson, 1983) and Yakutsk (Mikhailov and Pravdin, 1997), the possible anisotropy is equal to: (1.52±0.44)% and (1.35±0.36)%, respectively. However, the anisotropy phases differ by 90º (212º±17º and 123º), so the results should be treated with caution. At an energy of about 10 18 eV, the angular distribution of EASs in the AGASA experiment (Hayashida et al, 1999) correlates with the Galactic center (anisotropy ~4%), while at higher energies (>410 19 eV) the anisotropy disappears.
To select from possible origin models, information on the mass composition is also important. The available results are very uncertain. At energies of 10 17 –3 × 10 17 eV, according to data from the EAS installations at Moscow State University (Khristiansen et al, 1994) and Fly’s Eye (Bird et al, 1993), an enrichment of cosmic rays with heavy nuclei is observed, due to a break in the cosmic ray spectrum at an energy of ~3.10 15 eV. At energies above 1018 eV (Abbasi et al, 2005) and above 1019 eV (Shinozaki et al, 2003), the data do not contradict the assumption of the proton composition of CR.
Moving on to extremely high energies, we note the apparently established fact of the existence in CR of particles with an energy of more than 10 20 eV, which is significantly higher than the cutoff of the spectrum due to the GZK effect (Greisen, 1966; Zatsepin and Kuzmin, 1966), caused by the interaction of CR with relic photons. To date, according to various estimates, from 10 to 20 events have been recorded, with the maximum energy being ~3.10 20 eV.
To resolve the GZK paradox, various ideas have been proposed, which will be discussed in the section “Origin of CL”. Here we note one of the hypotheses associated with the possible violation of Lorentz invariance at ultra-high energies (Kirzhnits and Chechin, 1971), within which (Coleman and Glashow, 1999) neutral and charged pions can be stable particles at energies above 1019 eV and be part of the primary KL.
4. PROPAGATION OF COSMIC RAYS IN THE GALAXY
4.1 Basic parameters of the interstellar medium
The main features of the interstellar medium are its nonstationarity and a wide variety of physical conditions (Astrophysics KL, 1990). Interstellar gas, whose mass is 5·10 9 M O, exists in several modifications. The hot gas formed as a result of supernova explosions is characterized by a density n≈3·10 -3 /cm3, a temperature T≈10 6 K and occupies a fraction of f≈0.2-0.8 in the galactic disk. In addition, there is a warm intercloud medium (n≈0.1 cm -3, T≈104 K, f≈0.2-0.8), clouds of atomic hydrogen (n≈40 cm -3, T≈100 K, f≈0.03), molecular clouds (n≈200 cm -3, T≈10 K, f≈3·10 -3). The average concentration of hydrogen nuclei in the galactic disk is ≈1 cm-3>.
Most of the interstellar gas of the Galaxy, like most young stars, is concentrated in the spiral arms of the Galaxy, the width of which in the galactic plane is several hundred parsecs. The masses of atomic and molecular hydrogen are approximately equal (~2·10 9 MO). Hot gas from the disk should also penetrate into the halo, where it may contain about a few percent of the total gas mass; the concentration of hydrogen nuclei in the halo is ≈0.01 / cm 3.
Observations made by various methods indicate the existence of noticeable random motions of the interstellar medium with a maximum scale of ≈100 pc. The total energy density associated with random motions is about 1 eV/cm -3 , i.e., comparable to the energy density of cosmic rays.
The distribution of supernovae in the Galaxy is also not uniform, and, in addition to individual supernovae, there are clusters of them. As a result of successive supernova explosions within an association of OB stars, giant hot cavities (superbubbles) with dimensions of 10 2 -10 3 pc and with a total released energy of the order of 10 54 erg appear. The frequency of such processes in the Galaxy is estimated as 10 -4 per year, and the lifetime of the cavity is ~10 7 years.
An increased level of turbulence should be expected in caverns, which provides additional opportunities for cosmic ray acceleration (Bykov and Toptygin, 1995).
The process of cosmic ray propagation in the Galaxy obviously depends on the structure of magnetic fields. The regular field lines lie in the galactic plane and approximately run along the spiral arms. The average amplitude of the field strength is (2-3)·10 -6 G. The random component of the Galactic magnetic field is characterized by a main scale of L≈100 pc and an amplitude exceeding the amplitude of the regular field, so that () 1/2 /B reg ≈(1-3). The spectrum of magnetic field inhomogeneities is currently unknown exactly; however, it cannot be ruled out that this spectrum, like the spectrum of gas inhomogeneities, is close to the Kolmogorov spectrum in the scale range from 10 12 cm to 100 pc. A magnetic field also exists in a halo, and in the literature there is no single point of view regarding its magnitude.
4.2 CR diffusion in galactic magnetic fields
We have already mentioned above that cosmic rays do not propagate in a straight line, but diffuse in the magnetic fields of the Galaxy. The experimentally observed ratio of the fluxes of light and medium nuclei is (for nuclei with energies above 2.5 GeV/nucleon) NL/NM=0.3±0.05, while the corresponding value for stars is 10 -6. Consequently, cosmic rays are extremely enriched in light nuclei, and since these nuclei are practically absent from the sources, they appear as a result of interactions of heavier nuclei. In order for this to happen, estimates show that an amount of matter x g = (5–10) g/cm2 must pass through the interstellar medium. This value should be compared with the amount of matter in the Galaxy passed in a straight line x og =ρ·R G ≈0.01 g/cm 2 . The ratio xg/xog≈103, which means the need for diffusion. At an energy of several GeV per nucleon, the lifetime of cosmic rays is ≈3.10 7 years and then decreases.
In addition, since the Solar System is located on the periphery of the Galaxy, in the absence of diffusion (or weak diffusion), the flux from the center of the Galaxy could significantly exceed the flux from the opposite direction. But data on the anisotropy of the cosmic ray flux indicate that the magnitude of the anisotropy up to energies of 10 14 eV remains small (Diffusion in a magnetic field has not a scalar, but a tensor character. Let Ni(E,r,t) be the concentration of nuclei of group i with energy E , at a distance r (measured, for example, from the center of the Galaxy) at time t. The diffusion equation satisfied by Ni(E,r,t) has the form
Where Di is the diffusion tensor, bi(E) describes the continuous energy losses of particles, Ti and Tk are the lifetimes of particles relative to inelastic interaction, Pki are fragmentation coefficients that specify the average number of nuclei of group i arising in inelastic interactions of nuclei of group k, Q(E ,r,t) – source function.
Let us consider the simplest case when it is possible to neglect nuclear interactions and continuous energy losses (the latter is almost always true for ultra-high-energy cosmic rays, while neglecting nuclear interactions in some cases is unacceptable, as, for example, when estimating the flux of group L nuclei). Under these conditions, stationary the diffusion equation for any group of nuclei has the form (Astrophysics KL, 1990):
The components Dij of the diffusion tensor are defined as follows:
Dij=(D II -D ⊥)bibj +D⊥δij+DAe ijn b n ,
where bi=B0i/B0 is the component of the unit magnetic field vector; D II , D ⊥ and DA are the coefficients of parallel, perpendicular and Hall diffusion, respectively, δij is the Kronecker symbol, e ijn is the absolute antisymmetric tensor, the index defining the group of nuclei is omitted.
In real conditions of our Galaxy, the most significant role is played by the diffusion coefficients D ⊥ and DA. Note that Hall diffusion “in another language” is the drift of particles in the large-scale regular magnetic field of the Galaxy (Ptuskin et al, 1993). At low energies, significantly lower than the energy of 3.10 15 eV, at which a break in the energy spectrum of the GCR is observed, D ⊥ dominates, and ordinary scalar diffusion takes place with the diffusion coefficient D=D ⊥ , where D ⊥ is defined as follows:
D ⊥ ~D ⊥0 (E/3 GeV)m, m=(0.1-0.2).
The Hall diffusion coefficient DA is proportional to the Larmor radius of the particle, i.e. DA~E.
Let us emphasize an important circumstance inherent in solutions to the diffusion equation: if the diffusion coefficient is a function of energy, then the energy spectrum of cosmic rays near the Earth I(E) will be different from their spectrum in sources Q(E), namely I(E ~Q(E)/ D(E).
Information regarding the energy dependence of the diffusion coefficient can be obtained by studying the anisotropy δ as a function of energy.
The available data on anisotropy in the energy range 10 12 –10 15 eV (see Fig. 16) is difficult to reconcile with the assumption that D (and, therefore, δ) increases with energy as E 0.6-0.7, which is required to obtain the observed experimentally the GCR spectrum from the spectrum obtained in the model of CR acceleration on the shock fronts of expanding supernova shells with . It is possible to slightly lower the requirements for the growth of D with energy (to D~E 0.3) by considering the process of additional acceleration of particles during their propagation in the Galaxy. At the same time, the D~E type dependence (0.6-0.7) does not contradict the results of studying the energy dependence of the L/M ratio at energies up to 10 11 eV/nucleon.
4.3 Influence of drift in the regular magnetic field of the Galaxy
An irregularity in the primary energy spectrum at E~3.10 15 eV (see Fig. 15) was discovered about 50 years ago, but the question of what causes this break has not yet been finally resolved. Therefore, it is possible to interpret the kink as a result of the propagation of cosmic rays in the Galaxy. Since the existence of a dependence of the diffusion coefficient on energy changes the spectrum of cosmic rays compared to the source, the desired result can be obtained if up to 3.10 15 eV D(E) weakly depends on E, and then this dependence increases. Since the value of DA is proportional to the Larmor radius of the particle, then, starting from a certain energy, the influence of Hall diffusion will dominate, and the propagation mode will change with a transition to a stronger dependence D(E). In this approach, it is possible to correctly reproduce the primary energy spectrum in the energy range up to 10 17 eV. At higher energies, the diffusion approximation becomes inadequate and it is necessary to use direct modeling of the motion of charged particles in the magnetic fields of the Galaxy.
In the region of relatively low energies (E≤10 11 eV), instead of the diffusion approximation, a homogeneous model (otherwise called the leaky box model), which is a simplified version of the diffusion one, is used (Astrophysics KL, 1990). In the homogeneous model, the second term of the diffusion equation is replaced by Ni(T)/T CR (hom), where the parameter T CR (hom) represents the characteristic time of cosmic ray exit from the Galaxy. It is believed that diffusion occurs quite quickly, and the concentration of cosmic rays in the Galaxy is generally constant.
A homogeneous model can be formally obtained as the limiting case of a diffusion model under the condition of weak leakage of particles from the system. Calculations within the framework of a homogeneous model turn out to be much simpler than the process of solving diffusion equations, which is the reason for its wide popularity, however, the use of a diffusion model is, of course, more preferable.
4.4 Fractal diffusion
In recent years, ideas have become widespread (Lagutin and Tyumantsev 2003) according to which diffusion in the Galaxy should be considered rather as diffusion in a fractal-type medium, rather than as “ordinary” diffusion in a medium with continuous parameters. The basis for this approach is the presence of inhomogeneities in the spatial distribution of matter and, consequently, magnetic fields in the Galaxy. It is extremely important that the mentioned inhomogeneities, which cause the chaotic motion of cosmic rays, are observed on different scales. All this stimulates the development of new approaches to the propagation of cosmic rays in the Galaxy. In particular, accepting the assumption that the distribution of heterogeneities is fractal in nature means that it is necessary to move from ordinary diffusion in a homogeneous or quasi-homogeneous medium to diffusion in a fractal-type medium (the so-called anomalous diffusion). The described approach is being successfully developed, however, until now, work in this direction has not led to the abandonment of the traditional mathematical apparatus.
5. ORIGIN OF COSMIC RAYS
If we keep in mind the entire energy range in which cosmic rays are observed, then, of course, it should be recognized that there is no complete theory of this issue. Even with regard to the origin of GCR, it is hardly possible at present to claim more than the creation of reasonable models that explain the most significant facts.
These should, first of all, include the energy density of cosmic rays (~10-12 erg/cm 3), as well as the power-law form of the GCR energy spectrum, which does not undergo any sharp changes up to an energy of ~3 10 15 eV, where the index of the differential energy spectrum of all particles changes from -2.7 to -3.1.
5.1 Supernova explosions as the main source of galactic cosmic rays
The requirements for the energy power of sources generating cosmic rays are very high, so that ordinary stars in the Galaxy cannot satisfy them (PCR power is about 3·10 40 erg/sec). However, such power can be obtained from supernova explosions (this idea was expressed about 50 years ago (Ginzburg and Syrovatsky, 1963)). If energy is released during an explosion, ~10 51 erg, and explosions occur with a frequency of 1 time in 30–100 years, then the power generated during supernova explosions is ~ 10 42 erg/cm 3 and only a few are sufficient to provide the required power of cosmic rays percent of flare energy.
The question of the formation of the experimentally observed energy spectrum of GCRs is far from trivial. It is necessary to transfer the macroscopic energy of the magnetized plasma (the expanding shell of an exploding supernova) to individual charged particles, while ensuring an energy distribution that differs significantly from the thermal one.
5.2 Standard model of CR acceleration by shock waves
The most likely mechanism for GCR acceleration to an energy of ~10 15 eV, and possibly higher, seems to be the following. The movement of the shell ejected during the explosion generates a shock wave in the surrounding interstellar medium. The diffusion propagation of charged particles captured in the acceleration process allows them to repeatedly cross the shock wave front (Krymsky, 1977). Each pair of successive intersections increases the energy of the particle in proportion to the energy already achieved (the mechanism proposed by Fermi), which leads to acceleration of the GCR. As the number of shock wave front intersections increases, so does the probability of leaving the acceleration region, so that the number of particles decreases as the energy increases in an approximately power-law manner. Acceleration turns out to be very effective, and the spectrum of accelerated particles is hard: ~E -2 up to ~Emax – the maximum achievable energy of accelerated particles.
Therefore, it is necessary to take into account the reverse effect of cosmic rays (the most significant role of protons, since heavier nuclei can be considered as small impurities) on the medium, leading to modification of the shock wave and to the appearance, in addition to the usual thermal front, of a smooth extended section, the so-called prefront. This modification, in turn, affects the spectrum of cosmic rays. Thus, in the general case, it is impossible to use an approximation when the reverse influence of cosmic rays on the medium is not taken into account, and it is necessary to use a self-consistent solution, the process of which has not yet been fully worked out (in the sense that, perhaps, all the necessary factors have not yet been fully taken into account) . A reflection of this circumstance is the almost continuous growth observed over the past 10 years in the theoretical estimate of the maximum achievable energy. Thus, in the work (Berezhko and Ksenofontov, 1999) the following estimate is given for the maximum achievable energy Emax:
Emax=5 10 14 Z (E SN /10 51 erg)1/2 (M ej /1.4 M O) -1/6 (N H /3 10 -3 cm -3)1/3 (B 0 /3 μG), eV,
Where Z is the charge of the accelerated particle, E SN is the flare energy, Mej is the mass of the ejected shell, N H is the concentration of hydrogen atoms, B 0 is the magnetic field strength. The agreement between the calculation results (Berezhko, 2001) and the experimental spectra (Shibata, 1995), as can be seen from Fig. 18, is quite good.
The above formula assumes the use of the Bohm limit for the diffusion coefficient D B =(1/3)?R L .c, where R L is the Larmor radius of the particle.
The validity of this traditional approximation, generally speaking, is not obvious and can be questioned. Note that in the approximation that does not take into account the inverse influence of cosmic rays on the shock wave, the Emax estimate is approximately an order of magnitude lower. The acceleration time reaches ~10 4 years, but its efficiency (understood as the possibility of generating particles with an energy close to Emax) decreases with time, so the time during which particles with the highest energies can be accelerated is ~10 3 years.
Fig. 18. CR intensity near the Earth as a function of kinetic energy. Curves – calculation, points – experimental data.
It also follows from the formula that by changing the characteristics of the flare (for example, the energy released during so-called Hypernova flares can significantly exceed 10 51 erg) and taking into account the distribution of flares over ESN, the Emax limit can be significantly increased. In addition, the shock wave can propagate not in the average interstellar medium, but in a medium modulated by the previously emitted stellar wind and characterized by a significantly higher magnetic field strength (as in Wolf-Rayet stars). Finally, taking into account the fact that the flow instability of accelerated particles in the prefront of the shock wave leads to the appearance of strong magnetodynamic turbulence, which also increases the maximum energy of the accelerated particles. As a result, it cannot be ruled out that the estimate can be increased to Emax~10 17 .Z eV.
The situation with the experimental detection of acceleration by shock waves now looks not quite certain. In particular, analysis of gamma-astronomy data shows that gamma-ray bursts of high (~1 TeV) energies are not always observed from nearby supernova remnants and, conversely, there are sources of high-energy gamma quanta that are not visible either in optical or X-ray ranges. Therefore, it is possible that the origin of GCRs is not due solely to supernova explosions.
It should be noted that the calculated spectrum of cosmic rays, up to the maximum achievable energy, turns out to be very hard (E -2), so that to compensate for the difference between the theoretical (-2) and experimental (-2.7) spectrum indicators, a significant softening of the energy spectrum during propagation is required cosmic rays from sources. Such a softening can be achieved if the diffusion coefficient D~E 0.7, but this assumption leads to an excessively strong GCR anisotropy at energies less than 1014 eV, which contradicts experimental data. Therefore, it seems more natural to have a dependence of the type D~E 0.3 (which approximately corresponds to the Kolmogorov spectrum of turbulence) and to take into account the additional acceleration of particles during the propagation process.
It can be stated that with the proper choice of injection parameters (a rigorous injection theory has not yet been created), which determine the number of injected particles and their speed, and taking into account the steepening of the GCR spectrum compared to the spectrum in the sources due to the dependence of the diffusion coefficient on energy, the theory of GCR acceleration on shocks waves makes it possible to well describe the energy spectra of protons and nuclei up to the energy corresponding to the break in the spectrum.
As noted above, explosions of vernovae can occur in associations of O- and B-stars, and in this case the explosions turn out to be correlated in time and space (the lifetime of associations is ~107 years, their number reaches several thousand and the frequency of explosions is estimated as 10 -5 – 10 -6 per year). The result is the formation of a cavity (superbubble) with hot low-density plasma and dimensions reaching hundreds of parsecs. In this cavern, random magnetic fields with scales L up to several parsecs and amplitudes B of tens of microgauss can be generated. At energies not exceeding Emax, acceleration is carried out by individual shock waves, and at energies exceeding Emax, acceleration is carried out by an ensemble of shock waves and magnetic fields existing in the cavern (Bykov and Toptygin, 1995). The acceleration model in supernova associations allows us to qualitatively explain the GCR spectrum in the energy range 1015–1018 eV. In this approach, the break in the GCR energy spectrum is interpreted as a change in the acceleration regime.
5.3 Other acceleration mechanisms
When discussing supernova explosions, it should be noted that GCR acceleration can take place not only in their expanding shells, but also during the evolution of the remnants of exploded stars. The source of energy in this case is the rotational energy of the neutron star, reaching (for a mass of 1.4.M O and a radius of 10 6 cm) a value of 2·10 50 erg/(T 10) 2, where T 10 is the rotation period in units of 10 milliseconds. Since the magnetic field on the surface of the star reaches 10 12 G, the neutron star must intensively lose energy to magnetic dipole radiation. However, since the natural frequency of the plasma in the vicinity of the star is much greater than the rotation frequency of the dipole, there will be no propagation of the electromagnetic wave, and the acceleration process will be carried out by a standing shock wave. The maximum energy is estimated to be ~(10 17 –10 18).Z eV, and the effective acceleration time is estimated to be approximately ~10 years (Gaisser, 1990).
If the neutron star is part of a binary system, then acceleration can also occur due to the accretion process - the flow of matter onto the surface of the neutron star; in this case, the acceleration of cosmic rays is provided by gravitational energy.
Due to the fact that the CR flux contains particles with energies exceeding 10 20 eV, it is necessary to consider the available possibilities for acceleration to such energies. The source of particles of such energies, for example, as noted in (Ptuskin, 1995), may be a first-order Fermi process, but occurring during a galaxy collision. Such an event can occur with a frequency of approximately 1 time in 5.10 8 years. The maximum achievable energy is estimated to be 3.10 19 .Z eV. The process of acceleration by shock waves in jets generated by active galactic nuclei leads to a similar assessment. Approximately the same amount is estimated in models related to the consideration of acceleration by shock waves caused by accretion in galaxy clusters.
The highest estimates can be obtained within the framework of the model of the cosmological origin of gamma-ray bursts. In this model, as a result of the merger of neutron stars or black holes, ultrarelativistic shock waves are generated that propagate in the environment with a Lorentz factor Г~10 3 . The energy of a proton at rest in the laboratory system, as a result of reflection from the shock wave front, will increase to the value Г 2 Мс 2. Thus, in just one cycle the energy can increase 10 6 times, and after two cycles it can reach 10 21 eV.
It should, however, be recognized that all estimates of this kind remain at a semi-qualitative level, and the issues of obtaining the required intensity and shape of the energy spectrum of ultra-high energy cosmic rays are still awaiting solution.
Soon after the discovery of cosmic microwave background radiation, Greisen (1966) in the USA and Zatsepin and Kuzmin (1966) in the USSR simultaneously came to the conclusion that the presence of cosmic microwave background radiation should most seriously affect the shape of the energy spectrum of extremely high-energy cosmic rays, namely: the following should occur the so-called relict (or blackbody) cutoff of the spectrum in the region of extremely high energies, also called the GZK effect. When discussing the problem of sources of particles with energies ≥5.10 19 eV, exceeding the threshold of the GZK effect, it is necessary to keep in mind that the distances from which particles of such energies can reach the Earth are apparently limited by the limits of the local Supercluster of galaxies.
Meanwhile, there are no galaxies in it that have any advantages over our Galaxy in terms of capabilities for accelerating cosmic rays. But even taking into account the limited distances to sources, there remain quite a lot of candidates for the role of sources of particles of extremely high energies.
Sources of extremely high-energy particles can form under two fundamentally different groups of scenarios (Nagano and Watson, 2000). The first group (bottom-up) is characterized by the presence of acceleration; Moreover, to overcome the limitation of distances to sources, new particles are sometimes considered that arise from ordinary ones, but do not experience losses leading to the appearance of the GZK limit. This group should also include models in which the existence of significant fluxes of particles with energies above the threshold of the GZK effect is associated with a hypothetical violation of Lorentz invariance. The second group (top-down) consists of scenarios that do not require acceleration, since in them CRs arise as a result of decays or annihilation of so-called topological defects (cosmic strings, monopoles, etc.) that arose in the first moments of the expansion of the Universe in connection with phase transitions corresponding to the separation of the strong interaction from the electroweak (at a temperature of 10 15 –10 16 GeV) and then the separation of the electromagnetic interaction from the weak (at a temperature of ~10 2 GeV).
6. CONCLUSION
GCR research, which has been ongoing for several decades, has not led, however, to closing the “blank spots” in this interesting area, although many issues have been successfully resolved. It can be stated, for example, that the accumulated information is quite sufficient to assess the contribution of GCRs to the radiation background in spacecraft orbits. However, as particle energy increases, the quality of information deteriorates. The insufficient luminosity of the installations used at high altitudes and in outer space does not allow studying the region 10 14 –10 15 eV by direct methods with sufficient statistics, not to mention moving into the energy region in which the break in the GCR spectrum occurs. The consequence of this situation is some instability of experimental data, which in the region above 10 12 eV after new experiments change the intensity estimates by 20–30%. Therefore, the immediate and urgent task remains the creation of equipment with large geometric factors, which would make it possible to study the fracture region using direct methods.
The use of indirect methods (primarily the study of EASs) has made it possible over the last decade to achieve certain progress in the study of the energy spectrum of GCRs, although the problem of model dependence of the results still exists. Currently, the question of obtaining spectra of individual groups of nuclei has begun to find an experimental solution. It is safe to assume that the upcoming launch of the LHC collider in 2007 will provide information on hadronic interactions up to an equivalent energy of ~10 17 eV and will significantly reduce the current uncertainty arising when extrapolating phenomenological models of hadronic interactions to the ultrahigh energy region.
EAS installations of the next generation should provide precision studies of the energy spectrum and composition of cosmic rays in the region 10 17 –10 19 eV. In this region, apparently, there is a transition from GCR to CR of extragalactic origin.
One can also hope that in the coming years the question of the existence of the GZK effect will be finally resolved, for which there are now serious indications.
Literature
Astrophysics of cosmic rays. Ed. V.L. Ginzburg. M.: Science. 1990. 528 p.
Bazilevskaya G.A., Makhmutov V.S., Svirzhevskaya A.K., Svirzhevsky N.S., Stozhkov Yu.I., Long-term measurements of cosmic rays in the Earth’s atmosphere, Izv. RAS. Ser. fiz., T.69, No. 6, P.835-837, 2005.
Baiburina S.A., Borisov A.S., Guseva Z.M. etc. Experiment “Pamir”. Interactions of ultra-high energy cosmic ray hadrons. M.: Science. 1984. Proceedings of the Lebedev Physical Institute. T.154. P.3-217. Berezhko E.G., Ksenofontov L.T., Composition of cosmic rays accelerated in supernova remnants, JETP, T.116, P.737-759, 1999.
Bugakov V.V., Belyakov S.A., Grigorov N.L., Gubin Yu.V., Kalinkin L.F., Kakhidze G.P., Rapoport I.D., Savenko I.A., Smirnov A. .V., Shiryaeva V.Ya., Shishkov P.P., Shesterikov V.F., Yakovlev B.M., Trigubov Yu.V., Principles of design of scientific equipment for studying high-energy cosmic rays on the Proton space station 4", Izv. USSR Academy of Sciences, Ser. fiz., T.34, pp.1818-1828, 1970.
Bykov A.M., Toptygin I.N., Spectra of high-energy particles generated in OB associations, Izv. RAS. Ser. Fiz., T.59, No. 4, P.162-165, 1995.
Ginzburg V.L., Syrovatsky S.I., Origin of cosmic rays, M.: USSR Academy of Sciences. 1963. 384 p.
Greisen K., Widespread air showers. In: “Physics of cosmic rays”. Ed. J. Wilson. M.: IL. 1958. T.3. No.7-141.
Zatsepin G.T., Kuzmin V.A., On the upper limit of the spectrum of cosmic rays, JETP Letters, T.4, P.114-116, 1966.
Zirakashvili V.N., Klepach E.G., Ptuskin V.S., Rogovaya S.I., Christiansen G.B., Chuvilgin L.G., Diffusion of high-energy cosmic rays in the Galaxy, Izv. Academy of Sciences of the USSR. Ser. fiz., T.55, pp.2049-2051, 1991.
Kirzhnits D.A., Chechin V.A., Cosmic rays and elementary length, JETP Letters, T.14, P.261-262, 1971.
Krymsky G.F., Regular mechanism of acceleration of charged particles at the shock wave front., DAN SSSR, T. 234, P.1306-1308, 1977.
Kulikov G.V., Christiansen G.B., On the spectrum of wide air showers by the number of particles, JETP, T.35, P.635-640, 1958.
Lagutin A.A., Tyumentsev A.G., Energy spectra of cosmic rays in a galactic medium of fractal type, Izv. RAS. Ser. fiz., T.67, No. 4, P.439-442, 2003.
Mikhailov A.A., Pravdin M.I., Search for anisotropy of ultra-high energy cosmic rays, JETP Letters, T.66, P.289-292, 1997.
Nymmik R.A., Semi-empirical dynamic model of particle fluxes of galactic cosmic rays. (ISO and 2005 versions), Model of the Cosmos, SINP MSU, 2006
Christiansen G.B., Kulikov G.V., Fomin Yu.A., Ultra-high energy cosmic radiation, M.: Atomizdat. 1975. 254 p.
Abbasi R.U., Abu-Zayyad T., Amman J.F., Archbold G.C., Bellido J.A., Belov K., Belz J.W., Bergman D.R., Cao Z., Clay R.W., Monocular measurement of the spectrum of UHE cosmic rays by the FADC detector of the HiRes experiment, Astropart.Phys., 23, pp.157-174, 2005.
Ambrosio M. for Macro Collaboration, Search for the sideral and solar diurnal modulations in the MACRO muon data set, Phys. Rev. D 67, 042002 (2003).
Asakimori K., Burnett T.H., Cherry M.L., Chevli K., Christ M.J., Dake S., Derrickson J.H., Fountain W.F., Fuki M., Gregory J.C., Hayashi T., Holynsky R., Iwai J., Iyono A., Johnson J., Kobayashi M., Lord J., Miyamura O., Moon K., H., Nilsen B.S., Oda H., Ogata T., Olson E.D., Parnell T.A., Roberts F.E., Sengupta K., Shiina T. , Strausz S.C., Sugitate T., Takahashi Y., Tominaga T., Watts J.W., Wefel J.P., Wilczynska B., Wilczynski H., Wilkes R.J., Wolter W., Yokomi H., and Zager E., Cosmic ray proton and helium spectra: results from the JACEE experiment, Astrophys. J., Vol.502, pp.278-283, 1998.
Berezhko E.G., Particle acceleration in supernova remnants, Inv. Rapp. Highlight papers. 27th ICRC, Hamburg, 2001, pp.226-233.
Bird D.J., Corbato S.C., Dai H.Y., Dawson B.R., Elbert J.W., Gaisser T.K., Huang M.H.A., Kieda D.B., Ko S., Larsen C.G., Loh E.C., Salamon M.H., Smith J.D., Sokolsky P., Sommers P., Stanev T. ., Tilav T., Tang J.k.k., Thomas S.B., The cosmic ray composition above 0.1 EeV, in Proc. 23th ICRC, Calgary, 1993, Vol.2, pp.38-42.
Bishop C.M., Neural networks for pattern recognition, Oxford Univ. press, 1995, 504 p.
Blobel V., Unfolding methods in high-energy physics experiments, CERN report 85-09, pp.88-127, 1985.
Bonino G., Castagnoli G. Cini, Cane D., Taricco C., Bhandari N., Solar modulation of galactic cosmic ray spectra since the Maunder minimum, in Proc. 27th ICRC, Hamburg, 2001, Vol.9, pp.3769-3772. Casaus J. on behalf of the AMS.02 Collaboration. Cosmic ray astrophysics with AMS.02, in Proc. 28th ICRC, Tsukuba, 2003, Vol.4, pp.2149-2152.
Cocconi G., About the most energetic cosmic rays, Astropart. Phys., Vol.4, pp.281-283, 1996.
Coleman S., Glashow S.L., High-energy tests of Lorentz invariance, Phys. Rev. D59, 116008 (1999)
Cronin J.W., Cosmic rays: the most energetic particles in the universe, Rev. Mod. Phys., Vol.71, pp.165-172, 1999.
D"Agostini G., A multidimensional unfolding method based on Bayes" theorem, Nucl. Instr. Meth., Vol.A362, pp.487-498, 1995.
Fukunaga K., Introduction to statistical pattern recognition, N-Y: Academic press, 1972, 592 p.
Gaisser T.K., Cosmic rays and particle physics, Cambridge Univer. press, 1990, 279 p.
Glasmacher M.A.K., Catanese M.A., Chantell M.C., Covault C.E., Cronin J.W., Fick B.E., Fortson L.F., Fowler J.W., Green K.D., Kieda D.B. et al, The cosmic ray energy spectrum between 1014 and 1016 eV, Astropart. Phys., Vol.10, pp.291-302, 1999.
Glushkov A.V., Egorova V.P., Ivanov S.P., Knurenko S.P., Kolosov V.A., Krasilnikov A.D., Makarov I.T., Mikhailov A.A., Olzoyev V.V., Pisarev V.V., Pravdin M.I., Sabourov A.V., Sleptsov I.E., Struchkov G.G., Energy spectrum of primary cosmic rays in the energy region of 1017 – 1020 eV by Yakutsk array data, in Proc. 28th ICRC, Tsukuba, 2003, Vol.1, pp.389-392.
Gold R., On iterative unfolding method for response matrices., Argonne National Lab. Report. ANL?6984., 1964. 39 p.
Greisen K., End to the cosmic-ray spectrum?, Phys. Rev. Lett., Vol.16, pp.748-750, 1966. Haungs A., Rebel H., Roth M., Energy spectrum and mass composition of high-energy cosmic rays, Rep. Prog. Phys., Vol.66, pp.1145-1206, 2003.
Hayashida N., Nagano M., Nishikawa D., Ohoka H., Sakaki N., Sasaki M., Takeda M., Teshima M., Torii R., Yamamoto T. et al. The anisotropy of cosmic ray arrival directions around 1018 eV, Astropart. Phys., Vol.10, pp.303-311, 1999.
Heber B., Galactic and anomalous cosmic rays in the heliosphere, Invited, Rapporteur, and Highlight papers. 27th ICRC, Hamburg, 2001, pp.118-135.
Heck D., Knapp J., Capdevielle J.N., Schatz G., Thouw T., Forschungszentrum Karlsruhe Report, FZKA 6019, 1998, 90 p.
Hillas A.M., Two interesting techniques for Monte-Carlo simulation of very high energy hadron cascades, in Proc.17th ICRC, Paris, 1981, Vol.8, pp.193-196.
Hillas A.M., Shower simulation: Lessons from MOCCA, Nucl. Phys. B (Proc. Suppl.), 52B, 29-42, 1997.
H?randel J.R., On the knee in the energy spectrum of cosmic rays, Astropart. Phys., Vol.19, pp.193-220, 2003.
Kalmykov N.N., Khristiansen G.B., Cosmic rays of superhigh and ultrahigh energies, J. Phys. G: Nucl. Part. Phys., Vol.21, pp.1279-1301, 1995.
Kalmykov N.N., Ostapchenko S.S., Pavlov A.I., Quark-gluon string model and EAS simulation problems at ultra-high energies, Nucl. Phys. B (Proc. Suppl.), 52B, 17-28, 1997.
Kempa J., Cosmic ray studies by the emulsion method. – Techniques and results, Nucl. Phys. B. (Proc. Suppl.), Vol.52B, pp.43-55, 1997.
Khristiansen G.B., Fomin Yu.A., Kalmykov N.N., Kulikov G.V., Motova M.V., Ostapchenko S.S., Sulakov V.P., Trubitsyn A.V., Primary cosmic ray mass composition at energies 1015 - 1017 eV as measured by the MSU EAS array, Astropart. Phys., Vol.2, pp.127-136, 1994.
Klages H.O. for the KASCADE Collaboration, The extensive air shower experiment KASCADE – first results, in Proc. 25th ICRC, Durban, 1997, Vol.8, pp.297-306.
Lloyd-Evans, Watson A.A., Anisotropy measurements above 1015 eV, Invited Talks. 8th Europ. CR Symp. Ed. N. Iucci et al. Bologna, 1983, pp.81-97.
Nagano M., Watson A.A., Observations and implications of the ultrahigh-energy cosmic rays, Rev. Mod. Phys., Vol.72, No.3, pp.689-732, 2000.
Parker E.N., The passage of energetic particles through interplanetary space, Planet. Space. Sci., Vol.13, pp.9-17, 1965.
Ptuskin V.S., Rogovaya S.T., Zirakashvili V.N., Chuvilgin L.G., Khristiansen G.B., Klepach E.G., Kulikov G.V., Diffusion and drift of very high energy cosmic rays in galactic magnetic fields, Astron.?Astroph., Vol.268, pp.726-735 , 1993.
Ptuskin V.S., Cosmic ray propagation in the Galaxy, Inv. Rapp. Highlight papers. 24th ICRC, Rome, 1995, pp.755-764.
Ptuskin V.S., Zirakashvili V.N., On the spectrum of high-energy cosmic rays produced by supernova remnants in the presence of strong cosmic-ray instability streaming and wave dissipation, Astron.? Astroph., Vol.429, pp.755-765, 2005.
Roth M., Antoni T., Apel W.D. et al for KASCADE Collab., Determination of primary energy and mass in the PeV region by Bayesian unfolding techniques, Nucl. Phys. B (Proc. Suppl.), Vol.122, pp.317-320, 2003.
Sakaki N., Chikawa M., Fukushima M., Hayashida N., Honda K., Inoue N., Kadota K., Kakimoto F., Kamata K., Kawaguchi S., Kawakami S., Kawasaki Y., Kawasumi N ., Kusano E., Mahrous A.M., Mase K., Mizobuchi S., Morizane Y., Nagano M., Ohoka H., Sakurai N., Sasaki M., Sasano M., Shinozaki K., Takeda M., Teshima M., Torii R., Tsushima I., Uchihori Y., Yamamoto T., Yoshida S., and Yoshii H., Cosmic ray energy spectrum above 3x1018 eV observed with AGASA, in Proc. 27th ICRC, Hamburg, 2001, Vol.1, pp.333-336.
Sciutto S.J., AIRES: A system for air shower simulations (Version 2.2.0), astro-ph/9911331 (216 pages) Shibata T., Cosmic ray spectrum and composition; direct observation, Inv. Rapp. Highlight papers. 24th ICRC, Rome, 1995, pp.713-736.
Shinozaki K., Chikawa M., Fukushima M., Hayashida N., Honda K., Inoue N., Kadota K., Kakimoto F., Kamata K., Kawaguchi S., Kawakami S., Kawasaki Y., Kawasumi N ., Maze K., Mizobuchi S., Nagano M., Ohoka H., Osone S., Sakaki N., Sakurai N., Sasaki M., Sasano M., Shimizu H.M., Takeda M., Teshima M., Torii R., Tsushima I., Uchihori Y., Yamamoto T., Yoshida S., and Yoshii H., Chemical Composition of Ultra-high Energy Cosmic Rays Observed by AGASA, in Proc. 28th ICRC, Tsukuba, 2003, Vol.1, pp.401-404.
Simpson J.A., The cosmic radiation: reviewing the present and future, in Proc. 25th ICRC, Durban, 1997, Vol.8, pp.4-23.
Wefel J.P., To higher energy: balloon and satellite investigations around the knee, J. Phys. G., Vol.29, pp.821-830, 2003.
Wiebel-Sooth B., Biermann P.L., Meyer H., Cosmic rays. VII. Individual element spectra: prediction and data, Astron.? Astroph., Vol.330, pp.389-398, 1998.
Go to other pages of the project "SiZiF"
For contact:[email protected]
received at SiZiF 05.10.06
