Modeling Self-Subtraction in Angular Differential Imaging: Application to the HD 32297 Debris Disk

Abstract

We present a new technique for forward-modeling self-subtraction of spatially extended emission in observations processed with angular differential imaging (ADI) algorithms. High-contrast direct imaging of circumstellar disks is limited by quasi-static speckle noise and ADI is commonly used to suppress those speckles. However, the application of ADI can result in self-subtraction of the disk signal due to the disk's finite spatial extent. This signal attenuation varies with radial separation and biases measurements of the disk's surface brightness, thereby compromising inferences regarding the physical processes responsible for the dust distribution. To compensate for this attenuation, we forward-model the disk structure and compute the form of the self-subtraction function at each separation. As a proof of concept, we apply our method to 1.6 and 2.2 micron Keck AO NIRC2 scattered-light observations of the HD 32297 debris disk reduced using a variant of the "locally optimized combination of images" (LOCI) algorithm. We are able to recover disk surface brightness that was otherwise lost to self-subtraction and produce simplified models of the brightness distribution as it appears with and without self-subtraction. From the latter models, we extract radial profiles for the disk's brightness, width, midplane position, and color that are unbiased by self-subtraction. Our analysis of these measurements indicates a break in the brightness profile power law at r~110 AU and a disk width that increases with separation from the star. We also verify disk curvature that displaces the midplane by up to 30 AU towards the northwest relative to a straight fiducial midplane.