In situ observations of energetic particles at the Earth's bow-shock attainable by the satellite missions have long created an opinion that electrons are most efficiently accelerated in a quasi-perpendicular shock geometry. However, shocks that deemed responsible for the production of cosmic ray electrons and their radiation from sources such as supernova remnants are much more powerful and larger than the Earth's bow-shock. Their remote observations suggest that electrons are accelerated very efficiently in the quasi-parallel shocks as well. In this paper we investigate the possibility that protons accelerated to high energies create sufficient wave turbulence necessary for the electron preheating and subsequent injection into the diffusive shock acceleration in a quasi-parallel shock geometry. An additional test-particle-electron population, meant as a low-density addition to the electron core-distribution the hybrid simulation operates on, is introduced. Its purpose is to investigate how these electrons are energized by the "hybrid" electromagnetic field. The reduced spatial dimensionality allowed us to dramatically increase the number of macro-ions per numerical cell and achieve the converged results for the velocity distributions of test-electrons. We discuss the electron preheating mechanisms which can make a significant part of thermal electrons accessible to the ion-driven waves observed in hybrid simulations. We find that the precursor wave field supplied by ions has considerable potential to preheat the electrons before they are shocked at the subshock. Our results indicate that a downstream thermal equilibration of the hot test-electrons and protons does not occur. Instead, the resulting electron-to-proton temperature ratio is a decreasing function of the shock Mach number, $M_\mathrm{A},$ tending to a saturation at high $M_\mathrm{A}.$

Cosmic ray (CR) spectra, both measured upon their arrival at the Earth's atmosphere and inferred from the emission in supernova remnants (SNRs), appear to be significantly steeper than the "standard" diffusive shock acceleration (DSA) theory predicts. Although the reconstruction of the primary spectra introduces an additional steepening due to propagation effects, there is a growing consensus in the CR community that these corrections fall short to explain the newest high-precision data. Using 2D hybrid simulations, we investigate a new mechanism that may steepen the spectrum during the acceleration in SNR shocks. Most of the DSA treatments are limited to homogeneous shock environments. To investigate whether inhomogeneity effects can produce the necessary extra steepening, we assume that the magnetic field changes its angle along the shock front. The rationale behind this approach is the strong dependence of the DSA efficiency upon the field angle, $\theta_\mathrm{Bn}$. Our results show that the variation of shock obliquity along its face results in a noticeable steepening of the DSA spectrum. Compared to simulations of quasi-parallel shocks, we observe an increase of the spectral index by $\Delta q=0.1-0.15$. Possible extrapolation of the limited simulation results to more realistic SNR conditions are briefly considered.

The direct measurements of cosmic rays (CRs), after correction for the propagation effects in the interstellar medium, indicate that their source spectra are likely to be significantly steeper than the canonical $E^{-2}$ spectrum predicted by the standard Diffusive Shock Acceleration (DSA) mechanism. The DSA has long been held responsible for the production of galactic CRs in supernova remnant (SNR) shocks. The $\gamma$-ray 'probes' of the acceleration spectra of CRs on-the-spot, inside of the SNRs, lead to the same conclusion. We show that the steep acceleration spectrum can be attributed to the $combination$ of (i) spherical expansion, (ii) tilting of the magnetic field along the shock surface and (iii) shock deceleration. Because of (i) and (ii), the DSA is efficient only on two ``polar caps'' of a spherical shock where the local magnetic field is within $\simeq45^{\circ}$ to its normal. The shock-produced spectrum observed edge-on steepens with the particle energy because the number of freshly accelerated particles with lower energies continually adds up to a growing acceleration region. We demonstrate the steepening effect by obtaining an exact self-similar solution for the particle acceleration at expanding shock surface with an arbitrary energy dependence of particle diffusivity $\kappa$. We show that its increase toward higher energy steepens the spectrum, which deeply contrasts with the standard DSA spectrum where $\kappa$ cancels out.

Precise measurements of galactic cosmic rays revealed a significant difference between the rigidity spectral indices of protons and helium ions. This finding is a notable contrast to the commonly accepted theoretical prediction that supernova remnant (SNR) shocks accelerate protons and helium ions with the same rigidity alike. Most of the earlier explanations for the "paradox" appealed to SNR environmental factors, such as inhomogeneous $p$/He mixes in the shock upstream medium, variable ionization states of He, or a multi-SNR origin of the observed spectra. The newest observations, however, are in tension with most of them. In this paper, we show by self-consistent hybrid simulations that such special conditions are not vital for the explanation of the cosmic ray rigidity spectra. In particular, our simulations prove that an SNR shock can modify the chemical composition of accelerated cosmic rays by preferentially extracting them from a homogeneous background plasma without additional, largely untestable assumptions. Our results confirm the earlier theoretical predictions of how the efficiency of injection depends on the shock Mach number $M.$ Its increase with the charge-to-mass ratio saturates at a level that grows with $M.$ We have convolved the time-dependent injection rates of protons and helium ions, obtained from the simulations, with a decreasing shock strength over the active life of SNRs. The integrated SNR rigidity spectrum for $p$/He ratio compares well with the AMS-02 and PAMELA data.

High precision spectrometry of galactic cosmic rays (CR) has revealed the lack of our understanding of how different CR elements are extracted from the supernova environments to be further accelerated in their shocks. Comparing the spectra of accelerated particles with different mass to charge ratios is a powerful tool for studying the physics of particle injection into the diffusive shock acceleration (DSA). Recent AMS-02 demonstration of the similarity of He/$p$, C/$p$, and O/$p$ rigidity spectra has provided new evidence that injection is a mass-to-charge dependent process. We performed hybrid simulations of collisionless shocks and analyzed a joint injection of $p$ and He$^{2+}$ in conjunction with upstream waves they generate. By implication, our results equally apply to C and O fully ionized ions, since they have similar mass to charge ratios. By convolving the time-dependent injection rates of $p$ and He, obtained from the simulations, with a decreasing shock strength over the active life of SNRs, we generate the integrated SNR spectra for $p$ and He. These spectra are consistent with the AMS-02 and Pamela data and earlier theoretical predictions. Our interpretation of the elemental anomaly is therefore intrinsic to collisionless shock mechanisms and does not require additional assumptions, such as the contributions from several different SNRs, their inhomogeneous environments or acceleration from grains.

Recent observations of galactic cosmic rays (CR) in the 1-500 GeV energy range have revealed striking deviations from what deemed "standard." The anomalies cut across hadronic and leptonic CRs. I discuss findings that challenge physical mechanisms long held responsible for the CR production in galactic supernova remnants (SNR). I also consider some new physics of particle acceleration in SNR shocks that is not part of conventional models but may explain the anomalies. However, a possible 20-30\% excess remains unaccounted for in the $e^{+}/e^{+}$ ratio over the range of a few 100 GeV. If not explained by future models, it suggests an additional source of positrons such as a dark matter decay/annihilation or pulsar contribution. Earlier efforts to explain both the $e^{+}/e^{-}$ and $p$/He anomalies with the "standard" models by adjusting the SNR environmental parameters and multiple sources are critically assessed.

Shortfalls in cosmic ray (CR) propagation models obscure the CR sources and acceleration mechanisms. This problem became particularly obvious after the Fermi, Pamela, and AMS-02 have discovered the electron/positron and $p/$He spectral anomalies. Most of the CR models use diffusive propagation that is inaccurate for weakly scattered energetic particles. So, some parts of the spectra affected by the heliospheric modulation, for example, cannot be interpreted. I discuss and adopt an exact solution of the Fokker-Planck equation arXiv1610.01584, which gives a complete description of a ballistic, diffusive and transdiffusive (intermediate between the first two) propagation regimes. I derive a simplified version of an exact Fokker-Planck propagator that can easily be employed in place of the Gaussian propagator, currently used in major Solar modulation and other CR transport models.

An analytic solution for a Fokker-Planck equation that describes propagation of energetic particles through a scattering medium is obtained. The solution is found in terms of an infinite series of mixed moments of particle distribution. The spatial dispersion of a particle cloud released at t=0 evolves through three phases, ballistic (t<Tc), transdiffusive (t~Tc) and diffusive (t>>Tc), where Tc is the collision time.The ballistic phase is characterized by a decelerating expansion of the initial point source in form of a "box" distribution with thickening walls. The next, transdiffusive phase is marked by the box walls thickened to its size and a noticeable slow down of expansion. Finally, the evolution enters the conventional diffusion phase.

The unexpected energy spectrum of the positron/electron ratio is interpreted astrophysically, with a possible exception of the 100-300 GeV range. The data indicate that this ratio, after a decline between $0.5-8$ GeV, rises steadily with a trend towards saturation at 200-400GeV. These observations (except for the trend) appear to be in conflict with the diffusive shock acceleration (DSA) mechanism, operating in a \emph{single} supernova remnant (SNR) shock. We argue that $e^{+}/e^{-}$ ratio can still be explained by the DSA if positrons are accelerated in a \emph{subset} of SNR shocks which: (i) propagate in clumpy gas media, and (ii) are modified by accelerated CR \emph{protons}. The protons penetrate into the dense gas clumps upstream to produce positrons and, \emph{charge the clumps positively}. The induced electric field expels positrons into the upstream plasma where they are shock-accelerated. Since the shock is modified, these positrons develop a harder spectrum than that of the CR electrons accelerated in other SNRs. Mixing these populations explains the increase in the $e^{+}/e^{-}$ ratio at $E>8$ GeV. It decreases at $E<8$ GeV because of a subshock weakening which also results from the shock modification. Contrary to the expelled positrons, most of the antiprotons, electrons, and heavier nuclei, are left unaccelerated inside the clumps. Scenarios for the 100-300 GeV AMS-02 fraction exceeding the model prediction, including, but not limited to, possible dark matter contribution, are also discussed.

Cosmic rays (CR), constrained by scattering on magnetic irregularities, are believed to propagate diffusively. But a well-known defect of diffusive approximation whereby some of the particles propagate unrealistically fast has directed interest towards an alternative CR transport model based on the "telegraph" equation. However, its derivations often lack rigor and transparency leading to inconsistent results. We apply the classic Chapman-Enskog method to the CR transport problem. We show that no "telegraph" (second order time derivative) term emerges in any order of a proper asymptotic expansion with systematically eliminated short time scales. Nevertheless, this term may formally be\emph{ converted} from the \emph{fourth} order hyper-diffusive term of the expansion. But, both the telegraph and hyperdiffusive terms may only be important for a short relaxation period associated with either strong pitch-angle anisotropy or spatial inhomogeneity of the initial CR distribution. Beyond this period the system evolves diffusively in both cases. The term conversion, that makes the telegraph and Chapman-Enskog approaches reasonably equivalent, is possible only after this relaxation period. During this period, the telegraph solution is argued to be unphysical. Unlike the hyperdiffusion correction, it is not uniformly valid and introduces implausible singular components to the solution. These dominate the solution during the relaxation period. As they are shown not to be inherent in the underlying scattering problem, we argue that the telegraph term is involuntarily acquired in an asymptotic reduction of the problem.

Recent efforts in cosmic ray (CR) confinement and transport theory are discussed. Three problems are addressed as being crucial for understanding the present day observations and their possible telltale signs of the CR origin. The first problem concerns CR behavior right after their release from a source, such as a supernova remnant (SNR). At this phase the CRs are confined near the source by self-emitted Alfven waves. The second is the problem of diffusive propagation of CRs through the turbulent ISM. This is a seemingly straightforward and long-resolved problem, but it remains controversial and reveals paradoxes. A resolution based on the Chapman-Enskog asymptotic CR transport analysis, that also includes magnetic focusing, is suggested. The third problem is about a puzzling sharp ($\sim10^{\circ}$) anisotropies in the CR arrival directions that might bear on important clues of their transport between the source and observer. The overarching goal is to improve our understanding of all aspects of the CR's source escape and ensuing propagation through the galaxy to the level at which their sources can be identified observationally.

Energetic nonthermal particles (cosmic rays, CRs) are accelerated in supernova remnants, relativistic jets and other astrophysical objects. The CR energy density is typically comparable with that of the thermal components and magnetic fields. In this review we discuss mechanisms of magnetic field amplification due to instabilities induced by CRs. We derive CR kinetic and magnetohydrodynamic equations that govern cosmic plasma systems comprising the thermal background plasma, comic rays and fluctuating magnetic fields to study CR-driven instabilities. Both resonant and non-resonant instabilities are reviewed, including the Bell short-wavelength instability, and the firehose instability. Special attention is paid to the longwavelength instabilities driven by the CR current and pressure gradient. The helicity production by the CR current-driven instabilities is discussed in connection with the dynamo mechanisms of cosmic magnetic field amplification.

In this review we discuss some observational aspects and theoretical models of astrophysical collisionless shocks in partly ionized plasma with the presence of non-thermal components. A specific feature of fast strong collisionless shocks is their ability to accelerate energetic particles that can modify the shock upstream flow and form the shock precursors. We discuss the effects of energetic particle acceleration and associated magnetic field amplification and decay in the extended shock precursors on the line and continuum multi-wavelength emission spectra of the shocks. Both Balmer-type and radiative astrophysical shocks are discussed in connection to supernova remnants interacting with partially neutral clouds. Quantitative models described in the review predict a number of observable line-like emission features that can be used to reveal the physical state of the matter in the shock precursors and the character of nonthermal processes in the shocks. Implications of recent progress of gamma-ray observations of supernova remnants in molecular clouds are highlighted.

Supernova remnants (SNRs), as the major contributors to the galactic cosmic rays (CR), are believed to maintain an average CR spectrum by diffusive shock acceleration (DSA) regardless of the way they release CRs into the interstellar medium (ISM). However, the interaction of the CRs with nearby gas clouds crucially depends on the release mechanism. We call into question two aspects of a popular paradigm of the CR injection into the ISM, according to which they passively and isotropically diffuse in the prescribed magnetic fluctuations as test particles. First, we treat the escaping CR and the Alfven waves excited by them on an equal footing. Second, we adopt field aligned CR escape outside the source, where the waves become weak. An exact analytic self-similar solution for a CR "cloud" released by a dimmed accelerator strongly deviates from the test-particle result. The CR diffusion coefficient $D_{NL}$ is strongly suppressed compared to its background ISM value $D_{ISM}$: $D_{NL}\sim D_{ISM}\exp(-\Pi)<< D_{ISM}$ for sufficiently high field-line-integrated CR partial pressure, $\Pi$. When $\Pi>>1$, the CRs drive Alfven waves efficiently enough to build a transport barrier that strongly reduces the leakage. The solution has a spectral break at $p=p_{br}$, where $p_{br}$ satisfies the following equation $D_{NL}(p_{br})\simeq z^{2}/t$.

The proof of cosmic ray (CR) origin in supernova remnants (SNR) must hinge on full consistency of the CR acceleration theory with the observations; direct proof is impossible because of the orbit stochasticity of CR particles. Recent observations of a number of galactic SNR strongly support the SNR-CR connection in general and the Fermi mechanism of CR acceleration, in particular. However, many SNR expand into weakly ionized dense gases, and so a significant revision of the mechanism is required to fit the data. We argue that strong ion-neutral collisions in the remnant surrounding lead to the steepening of the energy spectrum of accelerated particles by \emph{exactly one power}. The spectral break is caused by a partial evanescence of Alfven waves that confine particles to the accelerator. The gamma-ray spectrum generated in collisions of the accelerated protons with the ambient gas is also calculated. Using the recent Fermi spacecraft observation of the SNR W44 as an example, we demonstrate that the parent proton spectrum is a classical test particle power law $\propto E^{-2}$, steepening to $E^{-3}$ at $E_{br}\approx7GeV$.

The much-anticipated proof of cosmic ray (CR) acceleration in supernova remnants (SNR) must hinge on full consistency of acceleration theory with the observations; direct proof is impossible because of the orbit scrambling of CR particles. The PAMELA orbital telescope revealed deviation between helium and proton CR spectra deemed inconsistent with the theory, since the latter does not differentiate between elements of ultrarelativistic rigidity. By considering an initial (injection-) phase of the diffusive shock acceleration (DSA), where elemental similarity does not apply, we demonstrate that the spectral difference is, in fact, a unique signature of the DSA. Collisionless plasma SNR shocks inject more He2+ relative to protons when they are stronger and so produce harder helium spectra. The injection bias is due to Alfven waves driven by the more abundant protons, so the He2+ ions are harder to trap by these waves because of the larger gyroradii. By fitting the p/He ratio to the PAMELA data, we bolster the DSA-case for resolving the century-old mystery of CR origin.

In shock precursors populated by accelerated cosmic rays (CR), the CR return current instability is believed to significantly enhance the pre-shock perturbations of magnetic field. We have obtained fully-nonlinear exact ideal MHD solutions supported by the CR return current. The solutions occur as localized spikes of circularly polarized Alfven envelopes (solitons, or breathers). As the conventional (undriven) solitons, the obtained magnetic spikes propagate at a speed $C$ proportional to their amplitude, $C=C_{A}B_{{\rm max}}/\sqrt{2}B_{0}$. The sufficiently strong solitons run thus ahead of the main shock and stand in the precursor, being supported by the return current. This property of the nonlinear solutions is strikingly different from the linear theory that predicts non-propagating (that is, convected downstream) circularly polarized waves. The nonlinear solutions may come either in isolated pulses (solitons) or in soliton-trains (cnoidal waves). The morphological similarity of such quasi-periodic soliton chains with recently observed X-ray stripes in Tycho supernova remnant (SNR) is briefly discussed. The magnetic field amplification determined by the suggested saturation process is obtained as a function of decreasing SNR blast wave velocity during its evolution from the ejecta-dominated to the Sedov-Taylor stage.

A mechanism for proton acceleration to ~10^21eV is suggested. It may operate in accretion flows onto thin dark matter filaments of cosmic structure formation. The flow compresses the ambient magnetic field to strongly increase and align it with the filament. Particles begin the acceleration by the ExB drift with the accretion flow. The energy gain in the drift regime is limited by the conservation of the adiabatic invariant p_perp^2/B. Upon approaching the filament, the drift turns into the gyro-motion around the filament so that the particle moves parallel to the azimuthal electric field. In this 'betatron' regime the acceleration speeds up to rapidly reach the electrodynamic limit $cp_{max}=eBR$ for an accelerator with magnetic field $B$ and the orbit radius $R$ (Larmor radius). The periodic orbit becomes unstable and the particle slings out of the filament to the region of a weak (uncompressed) magnetic field, which terminates the acceleration. The mechanism requires pre-acceleration that is likely to occur in structure formation shocks upstream or nearby the filament accretion flow. Previous studies identify such shocks as efficient proton accelerators to a firm upper limit ~10^19.5 eV placed by the catastrophic photo-pion losses. The present mechanism combines explosive energy gain in its final (betatron) phase with prompt particle release from the region of strong magnetic field. It is this combination that allows protons to overcome both the photo-pion and the synchrotron-Compton losses and therefore attain energy 10^21 eV. A requirement on accelerator to reach a given E_max placed by the accelerator energy dissipation \propto E_{max}^{2}/Z_0 due to the finite vacuum impedance Z_0 is circumvented by the cyclic operation of the accelerator.

Recent observations of the supernova remnant W44 by the \emph{Fermi} spacecraft observatory strongly support the idea that the bulk of galactic cosmic rays is accelerated in such remnants by a Fermi mechanism, also known as diffusive shock acceleration. However, the W44 expands into weakly ionized dense gas, and so a significant revision of the mechanism is required. In this paper we provide the necessary modifications and demonstrate that strong ion-neutral collisions in the remnant surrounding lead to the steepening of the energy spectrum of accelerated particles by \emph{exactly one power}. The spectral break is caused by Alfven wave evanescence leading to the fractional particle losses. The gamma-ray spectrum generated in collisions of the accelerated protons with the ambient gas is also calculated and successfully fitted to the Fermi Observatory data. The parent proton spectrum is best represented by a classical test particle power law E^-2, steepening to E^-3 at E_br~7GeV due to deteriorated particle confinement.

Both the acceleration of cosmic rays (CR) in supernova remnant shocks and their subsequent propagation through the random magnetic field of the Galaxy deem to result in an almost isotropic CR spectrum. Yet the MILAGRO TeV observatory discovered a sharp ($\sim10^{\circ})$ arrival anisotropy of CR nuclei. We suggest a mechanism for producing a weak and narrow CR beam which operates en route to the observer. The key assumption is that CRs are scattered by a strongly anisotropic Alfven wave spectrum formed by the turbulent cascade across the local field direction. The strongest pitch-angle scattering occurs for particles moving almost precisely along the field line. Partly because this direction is also the direction of minimum of the large scale CR angular distribution, the enhanced scattering results in a weak but narrow particle excess. The width, the fractional excess and the maximum momentum of the beam are calculated from a systematic transport theory depending on a single scale $l$ which can be associated with the longest Alfven wave, efficiently scattering the beam. The best match to all the three characteristics of the beam is achieved at $l\sim1$pc. The distance to a possible source of the beam is estimated to be within a few 100pc. Possible approaches to determination of the scale $l$ from the characteristics of the source are discussed. Alternative scenarios of drawing the beam from the galactic CR background are considered. The beam related large scale anisotropic CR component is found to be energy independent which is also consistent with the observations.

Strong astrophysical shocks, diffusively accelerating cosmic rays (CR) ought to develop CR precursors. The length of such precursor $L_{p}$ is believed to be set by the ratio of the CR mean free path $\lambda$ to the shock speed, i.e., $L_{p}\sim c\lambda/V_{sh}\sim cr_{g}/V_{sh}$, which is formally independent of the CR pressure $P_{c}$. However, the X-ray observations of supernova remnant shocks suggest that the precursor scale may be significantly shorter than $L_{p}$ which would question the above estimate unless the magnetic field is strongly amplified and the gyroradius $r_{g}$ is strongly reduced over a short (unresolved) spatial scale. We argue that while the CR pressure builds up ahead of the shock, the acceleration enters into a strongly nonlinear phase in which an acoustic instability, driven by the CR pressure gradient, dominates other instabilities (at least in the case of low $\beta$ plasma). In this regime the precursor steepens into a strongly nonlinear front whose size scales with \emph{the CR pressure}as $L_{f}\sim L_{p}\cdot(L_{s}/L_{p})^{2}(P_{c}/P_{g})^{2}$, where $L_{s}$ is the scale of the developed acoustic turbulence, and $P_{c}/P_{g}$ is the ratio of CR to gas pressure. Since $L_{s}\ll L_{p}$, the precursor scale reduction may be strong in the case of even a moderate gas heating by the CRs through the acoustic and (possibly also) the other instabilities driven by the CRs.

A gas cloud near a supernova remnant (SNR) provides a target for pp-collisions leading to subsequent gamma-ray emission through neutral pion decay. The assumption of a power-law ambient spectrum of accelerated particles with index near -2 is usually built into models predicting the spectra of very-high energy (VHE) gamma-ray emission from SNRs. However, if the gas cloud is located at some distance from the SNR shock, this assumption is not necessarily correct. In this case, the particles which interact with the cloud are those leaking from the shock and their spectrum is approximately monoenergetic with the injection energy gradually decreasing as the SNR ages. In the GLAST energy range the gamma-ray spectrum resulting from particle interactions with the gas cloud will be flatter than expected, with the cutoff defined by the pion momentum distribution in the laboratory frame. We evaluate the flux of particles escaping from a SNR shock and apply the results to the VHE diffuse emission detected by the HESS at the Galactic centre.

We present a theory for the generation of mesoscale ($kr_{g}\ll 1$, where $r_{g}$ is the cosmic ray gyroradius) magnetic fields during diffusive shock acceleration. The decay or modulational instability of resonantly excited Alfven waves scattering off ambient density perturbations in the shock environment naturally generates larger scale fields. For a broad spectrum of perturbations, the physical mechanism of energy transfer is random refraction, represented by diffusion of Alfven wave packet in $k-$space. The scattering field can be produced directly by the decay instability or by the Drury instability, a hydrodynamic instability driven by the cosmic ray pressure gradient. This process is of interest to acceleration since it generates waves of longer wavelength, and so enables the confinement and acceleration of higher energy particles. This process also limits the intensity of resonantly generated turbulent magnetic field on $r_{g}$ scales.

We suggest a physical mechanism whereby the acceleration time of cosmic rays by shock waves can be significantly reduced. This creates the possibility of particle acceleration beyond the knee energy at ~10^15eV. The acceleration results from a nonlinear modification of the flow ahead of the shock supported by particles already accelerated to the knee momentum at p ~ p_*. The particles gain energy by bouncing off converging magnetic irregularities frozen into the flow in the shock precursor and not so much by re-crossing the shock itself. The acceleration rate is thus determined by the gradient of the flow velocity and turns out to be formally independent of the particle mean free path (m.f.p.). The velocity gradient is, in turn, set by the knee-particles at p ~ p_* as having the dominant contribution to the CR pressure. Since it is independent of the m.f.p., the acceleration rate of particles above the knee does not decrease with energy, unlike in the linear acceleration regime. The reason for the knee formation at p ~ p_* is that particles with $p > p_*$ are effectively confined to the shock precursor only while they are within limited domains in the momentum space, while other particles fall into ``loss-islands'', similar to the ``loss-cone'' of magnetic traps. This structure of the momentum space is due to the character of the scattering magnetic irregularities. They are formed by a train of shock waves that naturally emerge from unstably growing and steepening magnetosonic waves or as a result of acoustic instability of the CR precursor. These losses steepen the spectrum above the knee, which also prevents the shock width from increasing with the maximum particle energy.

The problem of the cosmic ray origin is discussed in connection with their acceleration in supernova remnant shocks. The diffusive shock acceleration mechanism is reviewed and its potential to accelerate particles to the maximum energy of (presumably) galactic cosmic rays ($10^{18}eV $) is considered. It is argued that to reach such energies, a strong magnetic field at scales larger than the particle gyroradius must be created as a result of the acceleration process, itself. One specific mechanism suggested here is based on the generation of Alfven wave at the gyroradius scale with a subsequent transfer to longer scales via interaction with strong acoustic turbulence in the shock precursor. The acoustic turbulence in turn, may be generated by Drury instability or by parametric instability of the Alfven waves.

Cosmic rays (CRs) are thought to be accelerated in SNRs. The most favorable situation for proving that the main, hadronic CR component is accelerated there is when CRs interact with dense gases, such as molecular clouds (MC) which surround the SN shock. Here, a new mechanism of spectrum formation in partially ionized gases near SNRs is proposed. Using an analytic model of nonlinear diffusive shock acceleration, we calculate the spectra of protons and estimate the resulting $\gamma$-ray emission occurring when the SNR shock approaches a MC. We show that the spectrum develops a break in the TeV range and that its GeV component is suppressed. These modifications to the standard theory occur because of the proximity of the partially ionized MC-gas and because of the physics of particle and Alfven wave propagation inside the gas. Possible applications of the new spectra to the recent CANGAROO and HESS observations of the SNR RX J1713.7-3946 are discussed.

The theory of shock acceleration predicts the maximum particle energy to be limited only by the acceleration time and the size (geometry) of the shock. This led to optimistic estimates for the galactic cosmic ray energy achievable in the SNR shocks. The estimates imply that the accelerated particles, while making no strong impact on the shock structure (test particle approach) are nevertheless scattered by strong self-generated Alfven waves (turbulent boost) needed to accelerate them quickly. We demonstrate that these two assumptions are in conflict when applied to SNRs of the age required for cosmic ray acceleration to the ``knee'' energy. We study the combined effect of acceleration nonlinearity (shock modification by accelerated particles) and wave generation on the acceleration process. We show that the refraction of self-generated waves resulting from the deceleration of the plasma flow by the pressure of energetic particles causes enhanced losses of these particles. This effect slows down the acceleration and changes the shape of particle spectrum near the cut-off. The implications for observations of TeV emission from SNR remnants are also discussed.

The linear theory of shock acceleration predicts the maximum particle energy to be limited only by the acceleration time and the size of the shock. We study the combined effect of acceleration nonlinearity (shock modification by accelerated particles, that must be present in strong astrophysical shocks) and propagation of Alfven waves that are responsible for particle confinement to the shock front. We show that wave refraction to larger wave numbers in the nonlinearly modified flow causes enhanced losses of particles in the momentum range p_max/R < p < p_max, where R>1 is the nonlinear pre-compression of the flow and p_max is a conventional maximum momentum, that could be reached if there was no refraction.

One of the main features of astrophysical shocks is their ability to accelerate particles to extremely high energies. The leading acceleration mechanism, the diffusive shock acceleration is reviewed. It is demonstrated that its efficiency critically depends on the injection of thermal plasma into acceleration which takes place at the subshock of the collisionless shock structure that, in turn, can be significantly smoothed by energetic particles. Furthermore, their inhomogeneous distribution provides free energy for MHD turbulence regulating the subshock strength and injection rate. Moreover, the MHD turbulence confines particles to the shock front controlling their maximum energy and bootstrapping acceleration. Therefore, the study of the MHD turbulence in a compressive plasma flow near a shock is a key to understanding of the entire process. The calculation of the injection rate became part of the collisionless shock theory. It is argued that the further progress in diffusive shock acceleration theory is impossible without a significant advance in these two areas of plasma physics.

There are two distinct regimes of the first order Fermi acceleration at shocks. The first is a linear (test particle) regime in which most of the shock energy goes into thermal and bulk motion of the plasma. The second is an efficient regime when it goes into accelerated particles. Although the transition region between them is narrow, we identify the factors that drive the system to a {\it self-organized critical state} between those two. Using an analytic solution, we determine this critical state and calculate the spectra and maximum energy of accelerated particles.

As a rule, the shock compression controls the spectrum of diffusively accelerated particles. We argue that this is not so if the backreaction of these particles on the shock structure is significant. We present a self-similar solution in which the accelerated particles change the flow structure near the shock so strongly that the total shock compression may become arbitrarily large. Despite this, the energy spectrum behind the shock is close to E^{-3/2} independently of anything at all.

A simplified model of particle transport at a quasiparallel one-dimensional collisionless shock is suggested. In this model the MHD-turbulence behind the shock is dominated by a circularly polarized, large amplitude Alfv\'en wave originated upstream from the turbulence excited by particles leaking from the downstream medium. It is argued that such a wave having significantly increased its magnetic field during the transmission through the shock interface can effectively trap thermal ions, regulating their leakage upstream. Together with a background turbulence this wave also plays a fundamental role in thermalization of the incoming ion flow. The spectrum of leaking particles and the amplitude of the wave excited by these particles are selfconsistently calculated. The injection rate into the first order Fermi acceleration based on this leakage mechanism is obtained and compared with computer simulations. The related problem of shock energy distribution between thermal and nonthermal components of the shocked plasma is discussed. The chemical composition of the leaking particles is studied.

The efficiency and uniqueness of the diffusive shock acceleration is studied on the basis of the novel kinetic solutions. These solutions obtained earlier (paper I, astro-ph/9707152) selfconsistently describe a strong coupling of cosmic rays with the gas flow. They show that the dependence of the acceleration efficiency upon physical parameters is critical in nature. In this paper we investigate a steady acceleration in the parameter space formed by the injection rate (\nu ), the upper cut-off momentum p_1 and the Mach number M while the flow compression R serves as an order parameter. We determine a manifold of all possible solutions in this parameter space.

The selfconsistent steady state solution for a strong shock, significantly modified by accelerated particles is obtained on the level of a kinetic description, assuming Bohm-type diffusion. The original problem that is commonly formulated in terms of the diffusion-convection equation for the distribution function of energetic particles, coupled with the thermal plasma through the momentum flux continuity equation, is reduced to a nonlinear integral equation in one variable. Its solution provides selfconsistently both the particle spectrum and the structure of the hydrodynamic flow. A critical system parameter governing the acceleration process is found to be $\Lambda = M^{-3/4}\Lambda_1 $, where $ \Lambda_1 =\eta p_1/mc $, with a suitably normalized injection rate $ \eta $, the Mach number M >> 1, and the cut-off momentum $ p_1 $. We particularly focus on an efficient solution, in which almost all the energy of the flow is converted into a few energetic particles. It was found that (i) for this efficient solution (or, equivalently, for multiple solutions) to exist, the parameter $ \zeta =\eta\sqrt{p_0 p_1}/mc $ must exceed a critical value $\zeta_{cr} \sim 1$ ($p_0$ is the injection momentum), (ii) the total shock compression ratio r increases with M and saturates at a level that scales as $ r \propto \Lambda_1 (iii) the downstream power-law spectrum has the universal index q=3.5 over a broad momentum range. (iv) completely smooth shock transitions do not appear in the steady state kinetic description.