Deprojecting Sersic Profiles for Arbitrary Triaxial Shapes: Robust Measures of Intrinsic and Projected Galaxy Sizes

We present the analytical framework for converting projected light distributions with a Sersic profile into three-dimensional light distributions for stellar systems of arbitrary triaxial shape. The main practical result is the definition of a simple yet robust measure of intrinsic galaxy size: the median radius r (med), defined as the radius of a sphere that contains 50% of the total luminosity or mass, that is, the median distance of a star to the galaxy center. We examine how r (med) depends on projected size measurements as a function of Sersic index and intrinsic axis ratios, and demonstrate its relative independence of these parameters. As an application we show that the projected semimajor axis length of the ellipse enclosing 50% of the light is an unbiased proxy for r (med), with small galaxy-to-galaxy scatter of similar to 10% (1 sigma), under the condition that the variation in triaxiality within the population is small. For galaxy populations with an unknown or a large range in triaxiality an unbiased proxy for r (med) is 1.3 x R ( e ), where R ( e ) is the circularized half-light radius, with galaxy-to-galaxy scatter of 20%-30% (1 sigma). We also describe how inclinations can be estimated for individual galaxies based on the measured projected shape and prior knowledge of the intrinsic shape distribution of the corresponding galaxy population. We make the numerical implementation of our calculations available.

Automated summary

This study presents an analytical framework for converting projected light distributions with a Sersic profile into three-dimensional light distributions for stellar systems of arbitrary triaxial shape, and defines a practical measure of intrinsic galaxy size. The median radius r (med) is identified as a robust indicator of galaxy size, and the semimajor axis length of the enclosing ellipse is proposed as an unbiased proxy for r (med). Additionally, a method for estimating inclinations of individual galaxies based on projected shape is discussed.