The Nine Lives of Cosmic Rays in Galaxies

Cosmic-ray astrophysics has advanced rapidly in recent years, and its impact on other astronomical disciplines has broadened. Many new experiments measuring these particles, both directly in the atmosphere or space and indirectly via γ rays and synchrotron radiation, have widened the range and quality of the information available on their origin, propagation, and interactions. The impact of low-energy cosmic rays on interstellar chemistry is a fast-developing topic, including the propagation of these particles into the clouds in which the chemistry occurs. Cosmic rays, via their γ -ray production, also provide a powerful way to probe the gas content of the interstellar medium. Substantial advances have been made in the observations and modelling of the interplay between cosmic rays and the interstellar medium. Focusing on energies up to 1 TeV, these interrelating aspects are covered at various levels of detail, giving a guide to the state of the subject. 199 A nn u. R ev . A st ro . A st ro ph ys . 2 01 5. 53 :1 99 -2 46 . D ow nl oa de d fr om w w w .a nn ua lr ev ie w s. or g A cc es s pr ov id ed b y U ni ve rs ity o f M ar yl an d C ol le ge P ar k on 0 8/ 24 /1 5. F or p er so na l u se o nl y. AA53CH06-Grenier ARI 29 July 2015 12:16 1. SCOPE OF THE REVIEW The term cosmic rays (CRs) refers generically to particles with energies starting at 1 MeV and continuing to around 1021 eV. The particles encompass nuclei (from protons to actinides), antiprotons, electrons, and positrons. Locally, the CR energy density is dominated by GeV protons and compares with the interstellar energy densities of the thermal gas, magnetic field, stellar radiation, and the cosmic microwave background. The Larmor radii of Galactic CRs range from 105 km at the lowest energies to 10−1 pc near 1015 eV, so that directional information on their sources is completely lost during propagation. For this reason, γ -ray and radio synchrotron observations are essential for studying the large-scale distribution and spectra of Galactic CRs, giving clues to their origin and propagation. There is good evidence that CRs are accelerated in Galactic sites up to the knee energy of∼3×1015 eV at which point the spectrum steepens; at much higher energies they almost certainly originate outside the Galaxy because the Larmor radii are of kiloparsec order above 1018 eV, and there is no sign of a directional modulation related to the Galactic disc. The transition energy from Galactic to extragalactic, however, is not known because the nuclear compositions of both Galactic and extragalactic CRs affect the spectrum in the transition region through the dependence of the Larmor radius onmass and charge. The transition is usually assumed to be in the 1015−17 eV band in which the CR spectrum steepens from index 2.7 to approximately 3.3 (in particle flux per momentum unit), which is consistent with the point of rapid escape from the Galaxy. The spectrum flattens again at higher energies, consistent with a contribution from extragalactic sources. In contrast, CR electrons and positrons are all certainly of Galactic origin because their energy losses by inverse Compton scattering on the cosmic microwave background limit their travel in intergalactic space to only a few kiloparsecs, even at GeV energies. The centenary of the 1912 discovery of CRs has passed, accompanied by commemorative celebrations (Ormes 2013) and reviews (Kotera & Olinto 2011, Blasi 2013, Zweibel 2013, Blandford et al. 2014, Slane et al. 2014), that highlighted the impressive historical progress on the continuing questions of their origin and acceleration in the Galaxy and beyond. The main acceleration site of Galactic CRs is thought to be the shocks of young supernova remnants. Particles traverse the shock many times owing to efficient, Bohm-like diffusion. They gain momentum at each pass, generating a power-law spectrum. The number of remnants and the long-term average supernova frequency in the Galaxy are sufficient to supply the observed CRs if a few percent of the explosive energy of a supernova is converted into CRs (e.g., Drury et al. 1994). Gamma-ray observations of supernova remnants at GeV energies (Fermi Large Area Telescope, Fermi-LAT; Astro-Rivelatore Gamma a Immagini Leggero, AGILE) and at TeV energies (High Energy Stereoscopic System, HESS; Major Atmospheric Gamma-ray Imaging Cherenkov Telescopes, MAGIC; Very Energetic Radiation Imaging Telescope Array System, VERITAS) allow direct tests of this scenario via the production of pion-decay, bremsstrahlung, and inverse-Compton γ rays by in situ CRs. The challenge is to separate and delineate the emission components from the CR nuclei and electrons, i.e., to probe how efficiently and how high in energy supernova shock waves can accelerate both types of particles. A hadronic emission signature has at last been found in the W44 and IC 443 remnants with Fermi-LAT and AGILE (Ackermann et al. 2013b, Cardillo et al. 2014), although it does not constitute firm evidence of the acceleration of CR nuclei until the measurements are extended to well below the pion peak energy at 70 MeV. In parallel, a significant paradigm shift has been the realization that CR sources other than supernova remnants are probably important. Pulsars and pulsar-wind nebulae are now receiving more attention as sources of primary electrons and positrons, triggered by observations of more high-energy positrons than expected from hadronic interactions of CRs with interstellar matter (see Section 2.4). 200 Grenier · Black · Strong A nn u. R ev . A st ro . A st ro ph ys . 2 01 5. 53 :1 99 -2 46 . D ow nl oa de d fr om w w w .a nn ua lr ev ie w s. or g A cc es s pr ov id ed b y U ni ve rs ity o f M ar yl an d C ol le ge P ar k on 0 8/ 24 /1 5. F or p er so na l u se o nl y. AA53CH06-Grenier ARI 29 July 2015 12:16 CR sources and acceleration mechanisms are not addressed in the present review. For recent coverage, we refer the reader to the insightful reviews quoted above, and to Castellina & Donato (2013) and Amato (2014). We focus instead on the numerous facets of CR lives and feedback in their host galaxies, which we attempt to capture and review here in their fascinating variety. In broad brushstrokes, Galactic CRs: 1. Differentially penetrate into astrospheres to probe their physics, as glimpsed in the heliosphere by the two Voyager spacecraft and by γ -ray observations toward the Sun. 2. Randomly diffuse through the turbulent interstellar magnetic fields, scrambling their paths to our detectors but transferring and ultimately depositing energy across kiloparsecs in the interstellar medium (ISM). CR nuclei and electrons leave 10% and 60% of their energy, respectively, before escaping into intergalactic space. Energy losses and interactions with matter and radiation fields leave collective traces of their long journeys through the Galactic disc and halo. 3. Produce spallation secondaries with elemental and isotopic abundances of diagnostic value in understanding the composition of accelerated matter and later propagation. 4. Drive magnetohydrodynamic (MHD) waves, which partly maintain interstellar turbulence. Conversely, CRs are subjected to wind-driven turbulence near stellar clusters, which may severely reduce their diffusion lengths in star-forming regions. 5. Drive large-scale interstellar flows, including possible galactic winds, fountains, and perhaps large-scale features, like the Fermi Bubbles. 6. Take away between 10% and 50% of the energy of the supernova shock waves that have accelerated them, thereby lowering remnant temperatures and moderating their feedback on the ISM. 7. Ionize and heat the dense interstellar gas, depositing ∼13 eV per interaction to keep the darkest ISM at temperatures near 10 K and electron fractions near 10−7. They drive a remarkably rich chemistry at low temperature. Reactive molecular ions thus provide remote tracers of low-energy CRs, with kinetic energies below a few MeV. 8. Cause interstellar clouds to glow brightly in γ rays and reveal their total gas content, irrespective of their chemical and thermodynamical state. 9. Interact with magnetic fields, probing their strength, as well as with soft radiation fields to trace starlight and dust radiation. We have highlighted nine topics above, hence the title of this review (for completeness, we mention that there are other related aspects not addressed here, such as the effect ofCRs on climate and weather), which aims to discuss recent findings on the numerous interactions of subTeV CRs with their interstellar habitat. Some of these findings were not known at the time of the previousCR article in Annual Reviews (Strong et al. 2007), which illustrates how lively this field of research is. New results include direct composition measurements from light to heavy nuclei; evidence of spectral hardening above a few hundred GeV and of spectral differences between the nuclei; the successful mapping of remote CRs in the Milky Way from their γ -ray emission, as seen by the Fermi-LAT telescope; the rapid developments in the use of chemical information to probe few-MeV CRs beyond the enormous barrier of the heliosphere and near supernova remnants; rapid progress in the use of CRs and their γ -ray emission to trace the total gas in the ISM; the recent realization of the potential confinement of young CRs in starburst regions; and the first observations of CR-driven γ rays in starburst galaxies. This review focusses on CR energies only up to 1 TeV because it is mainly concerned with the interstellar effects of CRs and their use as a tracer of the ISM, where only lower energies are relevant. www.annualreviews.org • Cosmic Rays and the ISM 201 A nn u. R ev . A st ro . A st ro ph ys . 2 01 5. 53 :1 99 -2 46 . D ow nl oa de d fr om w w w .a nn ua lr ev ie w s. or g A cc es s pr ov id ed b y U ni ve rs ity o f M ar yl an d C ol le ge P ar k on 0 8/ 24 /1 5. F or p er so na l u se o nl y. AA53CH06-Grenier ARI 29 July 2015 12:16 Table 1 Recent measurements of cosmic-ray composition and spectraa Instrument Species Energy per nucleon Reference