Search for gamma-ray emission from bow shocks of runaway stars

The mystery of the origin of cosmic rays has been tackled for more than hundred years and is still not solved. Cosmic rays are detected with energies spanning more than 10 orders of magnitude and reaching energies up to ~ 10$^{21}$eV, far higher than any man-made accelerator can reach. Different theories on the astrophysical objects and processes creating such highly energetic particles have been proposed. A very prominent explanation for a process producing highly energetic particles is shock acceleration. The observation of high-energy gamma rays from supernova remnants, some of them revealing a shell like structure, is clear evidence that particles are accelerated to ultrarelativistic energies in the shocks of these objects. The environments of supernova remnants are complex and challenge detailed modelling of the processes leading to high-energy gamma-ray emission. The study of shock acceleration at bow shocks, created by the supersonic movement of individual stars through the interstellar medium, offers a unique possibility to determine the physical properties of shocks in a less complex environment. The shocked medium is heated by the stellar and the shock excited radiation, leading to thermal infrared emission. 28 bow shocks have been discovered through their infrared emission. Nonthermal radiation in radio and X-ray wavelengths has been detected from two bow shocks, pointing to the existence of relativistic particles in these systems. Theoretical models of the emission processes predict high-energy and very high-energy emission at a flux level in reach of current instruments. This work presents the search for gamma-ray emission from bow shocks of runaway stars in the energy regime from 100 MeV to ~ 100 TeV. The search is performed with the large area telescope (LAT) on-board theFermi satellite and the H.E.S.S. telescopes located in the Khomas Highland in Namibia. The Fermi-LAT was launched in 2008 and is continuously scanning the sky since then. It detects photons with energies from 20 MeV to over 300 GeV and has an unprecedented sensitivity. The all-sky coverage allows us to study all 28 bow shocks of runaway stars listed in the E-BOSScatalogue of infrared bow shocks. No significant emission was detected from any of the objects, although predicted by several theoretical models describing the non-thermalemission of bow shocks of runaway stars. The H.E.S.S. experiment is the most sensitive system of imaging atmospheric Cherenkov telescopes. It detects photons from several tens of GeV to ~ 100 TeV. Seven of the bow shocks have been observed with H.E.S.S. and the data analysis is presented in this thesis. The analyses of the very-high energy data did not reveal significant emission from any of the sources either. This work presents the first systematic search for gamma-ray emission from bow shocks of runaway stars. For the first time Fermi-LAT data was specifically analysed to reveal emission from bow shocks of runaway stars. In the TeV regime no searches for emission from theses objects have been published so far, the study presented here is the first in this energy regime. The level of the gamma-ray emission from bow shocks of runaway stars is constrained by the calculated upper limits over six orders in magnitude in energy. The upper limits calculated for the bow shocks of runaway stars in the course of this work, constrain several models. For the best candidate, ζ Ophiuchi, the upper limitsin the Fermi-LAT energy range are lower than the predictions by a factor ~ 5. This challenges the assumptions made in this model and gives valuable input for further modelling approaches. The analyses were performed with the software packages provided by the H.E.S.S. and Fermi collaborations. The development of a unified analysis framework for gamma-raydata, namely GammaLib/ctools, is rapidly progressing within the CTA consortium. Recent implementations and cross-checks with current software frameworks are presentedin the Appendix