Star clusters as engines of galaxy evolution

Stars form and live in hierarchical structures such as star clusters on the smallest scales up to groups of galaxies on the largest scales. The role of young and massive star clusters in the evolution of the host galaxies is one of the most active fields of research in modern astronomy. The most massive stars are able to inject into the interstellar medium large amounts of energy and momentum through their radiation, their stellar winds and ultimately by exploding as supernovae. Also because of their clustered configuration, massive stars have an impact on the host galaxy with their energetic output, which is often referred to as a whole with the term feedback. Feedback processes are invoked by numerical simulations in order to reproduce galaxies with realistic properties. In the most extreme scenarios stellar feedback can launch galactic winds that may suppress the star formation and mix the gas content not only within the galaxy but also with the intergalactic medium. In this thesis I describe the scientific background of my doctoral studies and present my research project on the local starburst galaxy Haro 11 (z=0.021). Together with my supervisors, we use multi-band photometry from the Hubble Space Telescope for resolving the nuclear starburst into tens of star clusters with masses between 105 and 107 solar masses and ages younger than 20 Myr. We observe that star-formation is propagating across the galaxy from the eastern side to the western side. Additionally, we model aperture-matched ultraviolet and optical spectroscopy and derive the physical properties of the stellar populations in the three starburst knots (A, B, C). This also allow us to estimate the following stellar feedback quantities: photo-ionisation rate, energy of stellar winds and supernovae, and their power. We compare the above measurements with the information on the kinematics of the ionised gas within the three knots, extracted from optical spectroscopy. Combining our analysis with previous works, we find a multi-phase medium with maximum outflow velocities up to 400 km/s. The three knots have at different times all hosted powerful outflows, whose energetics is consistent with the quantified stellar feedback. This indicates that the stellar winds and SN explosions have driven the detected outflows. Our work is important not only for shedding light on the physics of feedback, but also because it combines and compares different methodologies (i.e. photometry and spectroscopy, ultraviolet and optical wavelengths). I conclude with an outlook on the upcoming projects of my doctoral studies, which will exploit the methodology that I developed during the first two years of my PhD and extend it to a whole sample of star-forming nearby galaxies. Such projects also connect with the science that will be enabled by the future astronomical observatories. The work on Haro 11 is the subject of a paper recently submitted to the journal MNRAS. A copy of the paper can be found at the end of this thesis