Ultra High Energy Cosmic Rays Zosia A. C. Krusberg Department of Physics University of Chicago 6 June 2007

1

The Cosmic Ray Spectrum

Cosmic rays were first discovered in 1912, when Victor Hess brought electroscopes into hot-air balloons flying around 5000 meters into the Earth’s atmosphere. Hess received the Nobel Prize for this discovery in 1938. The same year, Pierre Auger proved the existence of extensive air showers caused by primary particles with energies above 1015 eV by simultaneously observing the arrival of secondary particles with Geiger counters separated by several meters [1]. Since these early discoveries, the study of the cosmic-ray spectrum has proceeded, both in the search for the high-energy end of the spectrum and in the quest for understanding the physical processes leading to the observed energies and fluxes. Now, after almost a century of research, the origin of cosmic rays is to a large extent still an open question. We know that below energies of around 100 MeV, the solar wind shields protons coming from outside the solar system, so the Sun must be giving rise to the observed fluxes. Above that energy, the cosmic-ray spectrum exhibits little structure and is approximated by broken power laws (see Figure 1). Up until the so-called ”knee” around 4 × 1015 eV, the cosmic rays are believed to originate from within the Milky Way, typically by shock acceleration in supernova remnants. Then, the ”ankle” at 5 × 1018 eV is usually interpreted as a crossover between the galactic and extragalactic cosmic ray components. Though there is a certain amount of disagreement as to the exact location of this crossover, observations of cosmic rays above the ankle have been found to be isotropic, suggesting that at least this component is extragalactic: at such high energies, galactic magnetic fields are no longer able to isotropize the cosmic rays, so if this component originated from within the galaxy, we would observe a very strong anisotropy toward the direction of the galactic plane [2].

2 2.1

Prerequisite Physics The GZK Effect

Soon after the discovery of the cosmic microwave background (CMB) radiation in 1965 it was realized that protons with sufficiently high energies would interact inelastically with CMB photons to produce pions. This process results in a drastic suppression of the UHECR spectrum right around 1020 eV, where photopion production becomes kinematically permitted. The physical reason for the suppression is that at energies around 1019 eV, the loss length for protons propagating through the CMB is of the order of a few Gpc, so we are receiving particles from nearly the entire visible universe. However, at energies around 1020 eV, the loss length has been reduced to around 100 Mpc, so at this energy we are observing particles from only a small fraction of the universe (see Figure 2) [2]. 1

2.2

Deflection by Magnetic Fields

One of the difficulties in understanding the origin of cosmic rays is that for most of their energy range, cosmic rays are deflected by magnetic fields permeating interstellar and intergalactic space. Galactic magnetic fields are known to be around a few micro-Gauss within the galactic disk, and are expected to decay exponentially away from the disk. Intergalactic magnetic fields are observed in dense galaxy clusters, and on larger scales magnetic fields are known to be weaker than around 10 nano-Gauss. As a result of these deflections, the arrival directions of the cosmic rays do not point back to the original sources. At cosmic ray energies around 1020 eV, however, interstellar and intergalactic magnetic fields are unable to cause significant deflection to particle orbits, so pointing to cosmic ray sources becomes possible. Recent high-resolution simulations of large-scale structure formation in a ΛCDM universe have followed the evolution of magnetic fields from seed fields in galaxies and clusters. Simultaneously, cosmic-ray protons are propagated through a volume with a radius of 110 Mpc. These simulations demonstrate that the deflection from the source position to the arrival direction for protons with arrival energy 4 × 1019 eV can reach around 1 degree in the most dense regions, whereas for protons with arrival energies of 1020 eV, the deflections are less than 0.1 degree (see Figures 3 and 4). This is significantly smaller than the resolution of contemporary UHECR observatories. Consequently, at these high energies, cosmic-ray arrival directions do point back to the original sources and cosmic-ray astronomy becomes possible [3].

3

Theoretical Considerations

Current theories of the origin of UHECRs are broadly categorized into two distinct scenarios: “bottom-up” acceleration scenarios and “top-down” decay scenarios, each containing a variety of different models. In a sense, the two scenarios are the exact opposite of one another. In bottomup scenarios, protons are accelerated from lower energies to the requisite high energies in specific astrophysical environments. In top-down scenarios, on the other hand, the high-energy particles arise from the decay of certain sufficiently massive particles originating from physical processes in the early universe.

3.1

“Bottom-Up” Scenarios

The earliest proposals for astrophysical environments capable of generating UHECRs included objects with strong magnetic fields and large jet outflows, such as active galactic nuclei (AGN) and pulsars. AGN are a broad class of objects generally characterized by the presence of a central supermassive black hole that is fueled by an accretion flow. It was believed that cosmic rays could be accelerated in the central regions of these AGN or in the jets that emerge from the disk. However, because of the strong magnetic fields and large photon densities in the centers of these objects, synchrotron radiation and photopion production limit the maximum energy of cosmic rays emerging from AGN to between 1015 − 1016 eV. For the same reason, similar limits on maximum cosmic-ray energies appear in the case of pulsars [2]. Radio galaxies of type FRII have been proposed as promising sites for the production of UHECRs. Radio galaxies are a subcategory of AGN that emit strongly in radio bands due to the presence of magnetic fields; they are categorized into two types based on the morphology of the large-scale radio emission: FRI-type radio galaxies are the brightest in their centers, whereas FRII-type galaxies are the brightest at the outer lobe regions. The hot spots in these lobe regions are terminated by strong shocks at which particles may be accelerated diffusively. Because of the weaker magnetic fields and lower photon densities in the acceleration region, energy losses are expected to be significantly smaller than in the central regions of AGN, and maximum energies as high as 1021 eV are believed to be obtainable with optimistic parameter choices [4].

2

Dead quasars, the objects left behind when AGN have run out of fuel, have also been proposed as potential UHECR sources. Though there is a relative scarcity of extremely luminous quasars with black hole masses > 109 M at large redshifts, the local number of dead quasars is expected to be quite large. In fact, one recent study suggests that the minimum number of dead quasars within a 50 Mpc radius is around 40. Though dead quasars are manifestly underluminous, their supermassive black holes are likely to be sufficiently spun-up to serve as high-energy accelerators of individual particles. In this scenario, magnetic field lines threading the event horizon of the black holes generate an effective emf as large as 1021 V by virtue of the induced rotation. Furthermore, energy losses due to synchrotron radiation and photopion production are expected to be less important than in active galaxies, though some energy losses due to pair production in collisions with ambient photons are possible. In theory, then, proton energies could reach the requisite high energies; however, because our understanding of these objects is limited, no detailed calculations of the spectrum of UHECRs they generate exists. It has been proposed that though dead quasars are not bright sources in either the optical or radio bands, they may emit gamma rays due to synchrotron radiation, so a deeper understanding of these objects should be attainable with future gamma-ray observatories [5]. The two most physically favorable sources of UHECRs appear to be FRII-type radio galaxies and dead quasars, although a number of uncertainties prevent exact calculations of the cosmicray spectra they generate. However, because these objects lie at different cosmological distances — FRII-type radio galaxies are located at distances greater than 100 Mpc, whereas dead quasars are abundant within 50 Mpc — their produced cosmic ray spectra are expected to be quite different. Specifically, cosmic rays originating in radio galaxies are unlikely to survive with any significant flux to our observatories due to the GZK effect, whereas cosmic rays from dead quasars may extend beyond this cutoff. Consequently, observations of the region around 1020 eV are of tremendous importance in determining the origin of UHECRs.

3.2

“Top-Down” Scenarios

In top-down scenarios, the energetics problem is solved trivially from the beginning. Here, UHECRs originate in the decay of some supermassive X particle with a mass mX > 1020 eV. This way, no acceleration mechanism is necessary. There are a number of possibilities for the sources of these massive particles, including topological defects such as cosmic strings or magnetic monopoles produced during symmetrybreaking phase transitions in the early universe. Topological defects can become unstable and decompose into constituent fields — superheavy gauge and higgs bosons — which then decay to produce UHECRs. This could occur, for example, when two segments of cosmic string or a monopole and antimonopole touch each other. In most cases the problem with UHECR production from topological defects is not the maximum energy, but the fluxes: in all cases, the calculated cosmic-ray fluxes are many orders of magnitude smaller than what is observed. One very general reason for these low fluxes is the large distance between topological defects, which tends to be of the order of a Hubble radius. Alternatively, the X particles could be supermassive metastable relic particles. In this scenario, cold dark matter (CDM) has a small component of long-lived superheavy particles. These particles must have masses greater than 1021 eV and lifetimes comparable to or greater than the age of the universe. Like other forms of non-dissipative CDM, X-particles would accumulate in the halo of our galaxy, and would therefore produce an UHECR spectrum without a GZK cutoff and without appreciable anisotropy [6, 7].

3

4

Observational Considerations

The presently available UHECR data is obtained by two experiments — AGASA (Akeno Giant Airshower Array) and HiRes (High-Resolution Fly’s Eye) — that use two different techniques to detect the extensive air showers (EASs) caused by cosmic rays in the atmosphere. AGASA uses an array of ground detectors to sample the lateral distribution of the EAS when it reaches the ground, whereas HiRes uses a telescope to observe the fluorescence light produced by the shower while it propagates in the atmosphere [8].

4.1

The GZK Effect

With regards to the GZK effect, the two experiments report conflicting findings at energies greater than 1020 eV. AGASA claims to have observed a continuation of the spectrum beyond the expected cutoff in the form of 11 super-GZK events (see Figure 5). These findings argue against the notion of extragalactic UHECR sources, at least in the form of FRII-type radio galaxies. In contrast, the spectrum from the HiRes experiment indicates smaller fluxes past 1020 eV, which is consistent with a GZK feature. Unlike AGASA, HiRes reports only two events with energies above 1020 eV (see Figure 6). However, both experiments suffer from extremely low statistics in this energy region, and the discrepancy between the two experiments is only around 3σ. Taking into account the systematic errors from the low-energy region, this discrepancy is reduced to only around 2σ. Nonetheless, whether the GZK feature is present in these spectra or not, the fact remains that events with energies greater than 1020 eV have been measured several times by different experiments, and the origin of these particles needs to be understood [8, 9].

4.2

Arrival Direction Anisotropies

Clustering on small angular scales is a signature that UHECRs are accelerated in specific astrophysical sources; furthermore, the number of multiplets is an index of the density of the sources. If no clustering is observed, on the other hand, UHECRs are more likely to originate from particle decays within our galactic halo. Consequently, small-scale anisotropies are an extremely important tool in understanding the origin of UHECRs. As in the case of the GZK feature, the two experiments disagree here as well. On large scales both experiments observe that arrival directions are isotropic (see Figure 7). However, AGASA reports the presence of small-scale clustering in its set of events with energies above 4 × 1019 eV. Specifically, the AGASA collaboration observed six doublets and one triplet with angular separation less than 2.5 degrees in a set of around 70 events (see Figure 8). HiRes, on the other hand, observed no such clustering [8, 9].

5

Outlook

Though detecting both the GZK feature and small-scale anisotropies in arrival directions is of great importance in deepening our understanding of UHECRs, the two most extensive cosmic ray experiments to date report contradictory data on these features. Fortunately, the Pierre Auger Observatory is likely to resolve these inconsistencies. Auger combines the two observational techniques of AGASA and HiRes, and ten percent of the detected events will be so-called hybrid events that are detected simultaneously by the two methods. This way, discrepancies between the energy assignments of the two techniques will be resolved. Furthermore, Auger expects huge statistics of events at high energies, so the spectrum in the GZK region will no longer be dominated by statistical fluctuations. These high statistics will be even more important in studying the anisotropies of the event arrival directions, so great progress is expected in this field in coming years [8].

4

References [1] A. V. Olinto, astro-ph/0410685 [2] P. Blasi, Magnetic Fields in the Universe, E. M. de Gouveia Dal Pino, G. Lugones, and A. Lazarian, eds. (American Institute of Physics, 2005) [3] K. Dolag, D. Grasso, V. Springel, and I. Tkachev, JKAS, 37, 427 (2004) [4] J. P. Rachen and P. L. Biermann, A&A 272, 161 (1993) [5] E. Boldt and P. Ghosh, MNRAS 307, 491 (1999) [6] V. Berezinsky, hep-ph/9802351 [7] P. Bhattacharjee and G. Sigl, Physics Reports 327, 109 (2000) [8] D. De Marco, astro-ph/0609118 [9] D. R. Bergman and J. W. Belz, astro-ph/0704.3721 [10] M. Kachelreiss, D. V. Semikoz, and M. A. Tortola, hep-ph/0302161

5

Figure 1: The cosmic ray spectrum (from http://astroparticle.uchicago.edu).

6

Figure 2: Loss length vs. energy for high-energy cosmic-ray protons [2].

7

Figure 3: Angular deflections for arrival energies of 4 × 1019 eV [3].

Figure 4: Angular deflections for arrival energies of 1020 eV [3].

8

Figure 5: The AGASA GZK Spectrum [9].

9

Figure 6: The HiRes GZK Spectrum [10].

10

Figure 7: The HiRes skymap for events with energies greater than 1019 eV [9].

11

Figure 8: The AGASA skymap for events with energies greater than 4 × 1019 eV. Large blue circles indicate doublets, the purple circle indicates a triplet [9].

12