The present numerical investigation delves into the intricate interplay between Mach number (Ms), flow characteristics, and vorticity dynamics within a T106A low-pressure turbine (LPT) blade passage. The two-dimensional (2D) compressible Navier–Stokes equations are solved using a high-accuracy, dispersion relation preserving methodology, which is validated against benchmark direct numerical simulations. Four Ms ranging from 0.15 to 0.30 are computed in order to display the intricate response of compressibility on the separation-induced transition process. The emergence and evolution of unsteady separation bubbles along the suction surface of the T106A blade are explored, revealing a growing trend with Ms. The time-averaged boundary layer parameters evaluated along the suction surface display a delayed separation with a smaller streamwise extent with increasing Ms. However, an overall increase in the blade profile loss and a decrease in turbulent mixing are observed with increasing Ms, suggesting a detrimental effect on LPT performance. Applying the compressible enstrophy transport equation (CETE) to the flow in a T106A blade passage reveals that while a linear relationship exists between Ms and certain CETE budget terms, other terms have a nuanced dependency, which paves the way for future investigations into the role of compressibility on enstrophy dynamics.