David Guszejnov

Research

About star formation

Stars are the fundamental pieces of cosmic evolution and their formation is a key process that influences the evolution of galaxies, planets, and the development of life. Despite its importance star formation is not-yet understood. This is because of the wide range of complicated physical processes involved (e.g., gravity, turbulence, magnetic fields, radiation).
We know that stars form in dense clouds of molecular Hydrogen that collapses under its own gravity and form stars of various masses. My goal is to understand the fundamental questions of str formation like: What regulates star formation? What sets the masses of stars? Why are stars clustered? How many stars are born as binaries? How do stars affect the cloud they form in? How is star formation different in other galaxies?

Infrared emission map of the Orion A molecular cloud. The cloud exhibits a filamentary structure in which dense cores form that will later collapse to form stars (Source: ESA/Herschel).

Simulating star formation

I am a lead developer of the STARFORGE collaboration. The STARFORGE suite are the first ever radiation-hydro simulations that follow the evolution of a giant molecular cloud (GMC) throughout its lifetime while also resolving the formation of individual low-mass stars. Such a dynamic range is necessary to investigate how GMC-scale properties influence star formation and how the process may vary in different galactic environments. Such simulations are also necessary to understand how stellar feedback influences gas dynamics in the cloud (e.g., leading to cloud disruption) and how it affects future SF (e.g., changing the IMF or quenching SF). The modular nature of the simulation will allow me to explore the role each physical process plays. For more information and videos see the STARFORGE website and Youtube.

STARFORGE_zoom — Zoomed in surface density maps for a cloud at the beginning of star formation. The final image (top, right) shows the kinetic energy weighted surface density, as well as the local velocity field around a protostellar jet.

Surface density (left) and line-of-sight velocity dispersion (right) maps of a simulated 20000 solar mass star-forming cloud with radiative feedback.

Semi-analytic modeling

Analytical models of star formation are useful for understanding the principal mechanisms behind different observed phenomena but are often unable to make quantitative predictions. To study in detail how the different physical processes affect the masses and distribution of stars I developed the MISFIT (MInimalist Star Formation Including Turbulence) semi-analytical framework. Due to its semi-analytical nature MISFIT can simulate a much wider dynamic range than conventional hydro codes: it can follow the evolution of clouds from the scale of giant molecular clouds (10 pc) down to the scales of protostars (0.1 AU). This framework provides a way to explore different star formation models and parameters at a modest computational cost.