In order to better understand star formation and its dependence on the environment within galaxies, we have to trace the star formation on scales from individuals stars to larger scale structures. We can then build connections between young stars and the gas they form from on the fundamental scales of stellar associations and molecular clouds. We identify stellar associations in nearby galaxies by applying a watershed algorithm to point source catalogs from the PHANGS-HST (Physics at High Angular resolution in Nearby GalaxieS with the Hubble Space Telescope) Treasury Program. We have developed a new method that applies the watershed algorithm to systematically trace the hierarchical structure of young stars and identify young stellar associations over a range of physical scales. We have tested our methods on two galaxies that span the distance range in the PHANGS-HST sample: NGC 3351 (10Mpc), NGC 1566 (18Mpc), and use the algorithm to trace structures on scales from 8 to 64 pc. We characterize the properties of the resulting associations including sizes, SED-fit ages, masses, and reddening. I will present a first look at the correlation of our watershed-defined regions with ALMA CO cloud catalogs and show how the nearest neighbor distribution varies across the different galaxy environments.