The Stellar Content and Dynamics of Superbubbles in the Large Magellanic Cloud

Marion Siang-li Oey

The University of Arizona, 1995
Advisor: Robert C. Kennicutt, Jr.

The interaction between massive stars and the interstellar medium (ISM) is a fundamental process determining the structure and composition of the ISM. This work examines the stellar content and resulting dynamics of superbubbles in the Large Magellanic Cloud (LMC).

In work with P. Massey, I first show analytically that for 2 single-O star bubbles in M33, the evolution of wind power as the stars evolve is important in the bubble evolution. In a second prototype study, we find that the LMC superbubble DEM 152 shows evidence for sequential star formation, based on differing ages between the stars interior and exterior to the shell. We construct a numerical form of the standard Weaver et al. (1977) evolutionary model for wind-driven bubbles, and use the stellar census to compare the predicted shell evolution with the observed kinematics. There is a substantial discrepancy: the shell's observed expansion velocity is too large relative to its radius.

I then find that the color-magnitude diagrams of the associations within 7 LMC superbubbles and 5 classical H II regions are indistinguishable. The H-R diagrams, constructed with spectral types for 6 superbubble clusters, also appear similar to those in classical H II regions, implying that the shell formation timescale is shorter than the cluster evolutionary timescale. The stellar winds of the 1--2 most massive stars must therefore dominate the shell formation. The star-forming events for the superbubble associations are also no more extended in duration than that of other OB associations. The slopes of the initial mass functions appear normal.

Numerical modeling of the 6 superbubbles shows results falling into two distinct categories: ``high-velocity'' objects showing anomalous kinematics like DEM 152 and ``low-velocity'' objects which appear fairly consistent with the model. X-ray evidence suggests that the high-velocity objects have been accelerated by supernova remnant (SNR) impacts. Results for both categories imply an overestimate in the growth rate equivalent to an effective input power of up to an order of magnitude too large.