Image-Constrained Modeling with Hubble and Keck Images Reveals that OGLE-2012-BLG-0563Lb is a Jupiter-Mass planet Orbiting a K Dwarf

TL;DR

This study combines Hubble and Keck imaging to accurately determine the mass of a star and its Jupiter-mass planet from microlensing data, revealing systematic errors in ground-based photometry and emphasizing the importance of image-constrained modeling.

Contribution

The paper introduces an image-constrained modeling method that improves mass estimates in microlensing events by identifying systematic errors in photometry.

Findings

Host star mass is 0.8 solar masses, planet mass is 1.1 Jupiter masses.

Systematic errors in ground-based photometry affected previous mass estimates.

Image-constrained modeling can identify and correct photometric systematic errors.

Abstract

We present high angular resolution imaging from the {\sl Hubble Space Telescope} combined with adaptive optics imaging results from the {\sl Keck}-II telescope to determine the mass of the OGLE-2012-BLG-0563L host star and planet to be $M_{\rm host} = 0.801\pm 0.033M_\odot$ and $M_{\rm planet} = 1.116 \pm 0.087 M_{\rm Jupiter}$, respectively, located at a distance of $D_L = 5.46\pm 0.56\,$kpc. There is a close-wide degeneracy in the light curve models that indicates star-planet projected separation of $1.50\pm 0.16\,$AU for the close model and $8.41\pm 0.87\,$AU for the wide model. We used the image-constrained modeling method to analyze the light curve data with constraints from this high angular resolution image analysis. This revealed systematic errors in some of the ground-based light curve photometry that led to an estimate of the angular Einstein Radius, $\theta_E$, that was too large by a factor of $\sim 2$. The host star mass is a factor of 2.4 larger than the value presented in the \citet{fukui15} discovery paper. Although most systematic photometry errors seen in ground-based microlensing light curve photometry will not be repeated in data from the {\sl Roman Space Telescope}'s Galactic Bulge Time Domain Survey, we argue that image constrained modeling will be a valuable method to identify possible systematic errors in {\sl Roman} photometry.