Slovene Pierre Auger Observatory Site

Home | More About PAO

The "all particle" cosmic ray spectrum presents several interesting features, especially above 10¹⁴ eV. This differential energy spectrum falls very rapidly with energy: it follows a power law decreasing roughly as E^-2.7 up to a region called the "knee" (around 10¹⁵ eV) and somewhat steeper above. This first structure is followed by a second one which occurs at above 10¹⁸ eV (or 1 EeV) where the spectrum seems to flatten (a structure usually called the "ankle"). There the data become very scarce to finally completely (or, as we shall see, almost completely) vanish above 100 EeV.

A special attention should be given to the rates with which the cosmic rays arrive in the vicinity of the earth's atmosphere, in terms of unit aperture and unit time, since those figures have dramatic consequences on the experimental techniques and methods. Up to the "knee" region, the rates are sufficient to use more or less small sized detectors shipped aboard satellites or balloons. Between the "knee" and the "ankle", the incident intensities become very small so that only large aperture ground detectors (including optical devices with their small -around 10%- duty cycles) can be used to reach reasonable statistics. Above the ankle, and especially above the so-called Greisen-Zatsepin-Kuzmin (GZK) cutoff, the fluxes are so infinitesimal that only giant ground arrays can be expected to help accumulating data.

Many questions remain partially or totally open in the understanding of the cosmic ray phenomena in this energy region.

• Although the existence of the "knee" structure in the spectrum can be understood to some extent (e.g. by a reasonable limit to the energies that can be produced by charged particle acceleration in supernova blast-waves), no natural explanation can be found for the sharpness of this structure.
• Since the study of the chemical composition of the cosmic rays by ground detectors is delicate and model dependent, it is important that a future generation of experiments make the link between composition measurements by direct and indirect methods around the "knee" region.
• There is a general agreement on the fact that the conventional acceleration mechanisms have difficulties in explaining the observations made of UHE cosmic rays up to the end of the spectrum. Although many authors propose models which, by using extreme values for the free parameters, can reach very high energy acceleration values, to our knowledge no satisfactory theory has been proposed up to now which can give the answers to the whole set of experimental observations available beyond the GZK cutoff.

In the following, we'll concentrate on the questions left open by the data available above the GZK cutoff and on a possible way to provide answers to them.

For more than 30 years, ground based detectors have been observing cosmic rays with energies equal to or larger than 100 EeV. Let us call such particles super cosmic rays (SCR) for brevity. Although, because of about 20% uncertainty in the values of the reconstructed energies, such a distinction is somewhat arbitrary, about 10 SCR were detected during this period. One must emphasize the fact that the detectors which reported these observations are designed with various techniques: Volcano Ranch (scintillators), Haverah Park (water Cerenkov), Akeno (scintillators and muon detection), Yakutsk (scintillators and Cerenkov), Fly's Eye (atmospheric fluorescence). So there is no doubt that the observations are reliable and not artefacts due to some misleading modelisation of the air showers. Many articles have been written about the SCR, and especially the most energetic among them, a 320 EeV event observed on 15 October 1991 by the Fly's Eye at Dugway (USA). The difficulties in interpreting such events can be summarized in a few words:

• There are very few astrophysical mechanisms able to accelerate particles to energies exceeding 100 EeV. If the mechanism is an electromagnetic process, the energy of the particle is then proportional to the magnetic rigidity in the region where the acceleration takes place. This puts very strong constraints on the site where the mechanism, whatever it is, can take place. Hillas proposed an often used plot showing the necessary magnetic fields and corresponding site dimensions to produce particles accelerated up to 100 EeV. The sites passing the threshold values are scarce: pulsars with extreme magnetic field values, active galactic nuclei, radiogalaxy lobes. Alternative ways of producing the SCR exist, such as the disintegration of grand-unification gauge bosons of masses as large as 10¹⁵ GeV and resulting from the collapse of cosmic strings or other topological defects. Such hypotheses are of course highly speculative and have been criticized by some authors.
• Once produced, the particles have to escape from the production sites. The above mentioned conventional mechanisms all present a very hostile environment where a large part of the initial energy should be lost, either by synchrotron radiation or through interactions with very dense radiation.
• If one ignores the preceding limitations, the current particles (except neutrinos) interact with the 3K microwave background radiation if their energies are above photoproduction thresholds (typically 200 TeV for photons, 70 EeV for protons). The corresponding attenuation lengths are less than a few tens of Mpc: this is the origin of the GZK cutoff. Also, at energies around 100 EeV or beyond, the bending of the charged particle trajectories by the extragalactic magnetic fields is weak enough so that the direction of the incident SCR must point within a few degrees at the source. It is a fact that no known conventional high-energy source is to be found within, say, 100 Mpc of our galaxy and in the directions of the incident SCR.

There have been many attempts to interpret the experimental data with the constraints listed above. The main conclusions one can reach in going through them is the following. There is no satisfactory way to explain the bulk of existing data with conventional physics. All the other possible explanations either lead to new interpretations of what is thought to be well known (conventional mechanisms but not yet observed or discovered, complete revision of our present knowledge about the extragalactic magnetic fields etc...) or to new physics (diffuse sources such as cosmic strings, neutrino interactions inconsistent with the Standard Model etc...).

There is no doubt that the observation of the SCR raises some of the most exciting problems to be solved in the field of high-energy astrophysics. The answers can only come from a dedicated experiment which should provide enough statistics of high-quality data so that we have a better understanding of the chemical composition of these objects, of their incoming direction and of the shape of their energy spectrum beyond the GZK cutoff. The main constraint on such an experiment is the expected incident flux of these cosmic rays. A back-of-the-envelope formula on this flux is:

I(E>E₀)=100/E₀²

which gives their intensity in km^-2.sr^-1.yr^-1 above the energy E₀ (in EeV). Above the GZK cutoff the rate is roughly 10 events per century and per square kilometer for a standard ground detector. Thus a 600 events per year experiment needs a 6000 km² ground array efficient 100% of the time. This is the starting point of the international giant array project presently known as the "Pierre Auger Observatory".

The design of the proposed detector is now finished and a final technical proposal is written. The present status of the project is that of a hybrid detector which combines the statistical power of a giant array (6000 km² installed over two sites with about 3200 water Cerenkov tanks) with the observational quality of a fluorescence telescope based on the Fly's Eye technique. The detector should use unconventional technologies such as solar power, data transfer and acquisition with cellular telephony or some other telecommunications standard, synchronisation through GPS satellites etc... It is designed in such a way that the identification of the incident particles should be possible, at least to some extent (distinction between photons, protons and heavy nuclei). It is hoped that in a very few years, the Auger Observatory will help to disentangle the threads of this astrophysical enigma.

Updated: November 7, 2006