Massive stars blow powerful stellar winds throughout their evolutionary stages from the main sequence to Wolf-Rayet phases. The amount of mechanical energy deposited in the interstellar medium by the wind from a massive star can be comparable to the explosion energy of a core-collapse supernova that detonates at the end of its life. In this study, we estimate the kinetic energy deposition by massive stars in our Galaxy by considering the integrated Galactic initial mass function and modeling the stellar wind luminosity. The mass loss rate and terminal velocity of stellar winds during the main sequence, red supergiant, and Wolf-Rayet stages are estimated by adopting theoretical calculations and observational data published in the literature. We find that the total stellar wind luminosity due to all massive stars in the Galaxy is about Lw ≈ 1.1 × 1041 erg s−1, which is about 1/4 of the power of supernova explosions, LSN ≈ 4.8 × 1041 erg s−1. If we assume that ∼ 1 − 10 % of the wind luminosity could be converted to Galactic cosmic rays (GCRs) through collisonless shocks such as termination shocks in stellar bubbles and superbubbles, colliding-wind shocks in binaries, and bow-shocks of massive runaway stars, stellar winds might be expected to make a significant contribution to GCR production, though lower than that of supernova remnants.

The galactic magnetic field (GMF) and the intergalactic magnetic field (IGMF) affect the propagation of ultra-high energy cosmic rays (UHECRs) from the source to us. Here we examine the in uences of the GMF/IGFM and the dependence of their sky distribution on galactic latitude, b. We analyze the correlation between the arrival direction (AD) of UHECRs observed by the Pierre Auger Observatory and the large-scale structure of the universe in regions of sky divided by b. Specifically, we compare the AD distribution of observed UHECRs to that of mock UHECRs generated from a source model constructed with active galactic nuclei. Our source model has the smearing angle as a free parameter that re ects the de ection angle of UHECRs from the source. The results show that larger smearing angles are required for the observed distribution of UHECRs in lower galactic latitude regions. We obtain, for instance, a 1σ credible interval for smearing angle of 0° ≤ θs ≤ 72° at high galactic latitudes, 60° < b ≤ 90°, and of 75° ≤ θs ≤ 180°, -30° ≤ b ≤ 30°, at low galactic latitudes, respectively. The results show that the in uence of the GMF is stronger than that of the IGMF. In addition, we can estimate the strength of GMFs by these values; if we assume that UHECRs would have heavier nuclei, the estimated strengths of GMF are consistent with the observational value of a few μG. More data from the future experiments may make UHECR astronomy possible.