The observation of the diffuse Galactic gamma ray flux is the most powerful tool to study cosmic rays in different regions of the Galaxy , because the energy and angular distributions of the photons encode information about the density and spectral shape of relativistic particles in the entire Milky Way .
An open problem of fundamental importance is whether cosmic rays in distant regions of the Milky Way have the same spectral shape observed at the Earth or not .
If the spectral shape of protons and nuclei is equal in all the Galaxy , the dominant , hadronic component of the diffuse gamma ray flux must have an angular distribution that , after correcting for absorption effects , is energy independent .
To study experimentally the validity of this factorization of the energy and angular dependence of the diffuse flux it is necessary to compare observations in a very broad energy range .
The extension of the observations to energies E _ { \gamma } \simeq 0.1 –10 PeV is of great interest , because it allows the study of the cosmic ray spectra around the feature known as the “ knee ” .
The absorption probability for photons in this energy range is not negligible , and distorts the energy and angular distributions of the diffuse flux , therefore a precise calculation of the absorption effects is necessary for the interpretation of the data .
In this work we present predictions of the diffuse gamma ray flux at very high energy , constructed under different hypothesis for the space dependence of the cosmic ray energy spectra , and discuss the potential of the observations for present and future detectors .