The First 4 Years of SN 1993J Revisited: Geometric $m$-ring Modeling of the Radio Shell with Closure Quantities Only

SN 1993J is the best-studied radio supernova, with observations using very-long-baseline interferometry (VLBI) spanning from within weeks of explosion through nearly three decades of ejecta evolution. Imaging and modeling techniques have revealed an expanding shell-like ejecta structure, with a width well-constrained after ~1000 days. In this work, we present a re-analysis of the first ~1700 days of SN 1993J evolution, using a new VLBI analysis technique with somewhat intrinsically higher angular resolution and compare our results with those from previous conventional techniques. We adopt the m-ring model, with potentially somewhat higher angular resolution, but only as a phenomenological alternative to the physically-motivated spherical shell model used in past analyses and shape the profile to approximately match that of the shell. We use closure quantities only, providing calibration insensitive constraints on ejecta geometry and largely independent comparisons with previous results from the same data sets. Using this approach we find the ejecta expanding with a power-law exponent $\omega=0.80\pm0.01$ averaged for the time from 175 to 1693 days, consistent with shell-fitting analyses. For the first time, we report estimates of the ejecta width at t<1000 days -- as early as 264 days post-explosion -- finding a mean fractional width with standard deviation of $0.24 \pm 0.04$ (of the radius) with no significant evolution. We present a fit of the azimuthal brightness modulation over time with the maximum of a horseshoe pattern rotating from east to south-southwest from 175 to 1000 d and then evolving to a more complex behavior, quantifying previous results from images only. The constraints on the angular brightness modulation and width over time and ~700 days earlier than existing analyses provide an avenue to test more advanced simulations of the explosion and interaction mechanisms.