### Abstract

Contributed Talk - Splinter ISM

Thursday, 24 September 2020, 16:32 (virtual room F)

## The blunt body problem in the context of stellar bow shocks

P.Knospe, M.M. Schulreich

Center for Astronomy and Astrophysics, Berlin Institute of Technology

The Interstellar Medium (ISM) can be observed in almost all wavelengths. However, determining its actual physical properties is rather difficult. One possibility to determine them is via shock phenomena. Bow shocks are ubiquitous in the universe and occur on all scales from the Earthâ€™s bow shock up to galactic scales. In fact, whenever a blunt body plunges through a compressible medium faster than the local speed of sound, the generation of a bow shock wave is inevitable and hence we investigate bow shocks generated by runaway stars. Unfortunately, this so-called blunt body problem is mathematically extremely challenging, as the type of the governing partial differential equations changes from elliptic to hyperbolic at the subsonic-supersonic flow transition. Therefore, analytic results are scarce and can only be found under restricting assumptions. We employ an inverse method, where the shock is assumed to be strong and stationary. It is parameterized in order to calculate the shape of the shock-generating body, the standoff distance, and the hydrodynamic variables in the shock layer. When expressed in a streamline coordinate system, one finds expressions that only depend on boundary conditions and the equation of state. Those can then be integrated numerically and converted back into a spatial coordinate system. We compare the results not only to observations but also to full hydrodynamic simulations. Two-fluid simulations from the literature, where the involved gas and dust are treated as coupled fluids, show that the analytic method only holds for the gas component, which has to be kept in mind for comparison with observations. Furthermore we see, that the shock-generating body is not the star itself, but its wind shell, which has to be fully developed for the problem to be stationary. Hence we run 3d simulations, using a grid based hydro solver with adaptive mesh refinement (RAMSES) to investigate, e.g., the timescale in which the wind becomes stationary.