Identification of known objects in solar system surveys

The discovery of new objects in modern wide-field asteroid and comet surveys can be enhanced by first identifying observations belonging to known solar system objects. The assignation of new observations to a known object is an attribution problem that occurs when a least squares orbit already exists for the object but a separate fit is not possible to just the set of new observations. In this work we explore the strongly asymmetric attribution problem in which the existing least squares orbit is very well constrained and the new data are sparse. We describe an attribution algorithm that introduces new quality control metrics in the presence of strong biases in the astrometric residuals. The main biases arise from the stellar catalogs used in the reduction of asteroid observations and we show that a simple debiasing with measured regional catalog biases significantly improves the results. We tested the attribution algorithm using data from the PS1 survey that used the 2MASS star catalog for the astrometric reduction. We found small but statistically significant biases in the data of up to 0.1 arcsec that are relevant only when the observations reach the level of accuracy made possible by instruments like PS1. The false attribution rate was measured to be < 1/1,000 with a simple additional condition that can reduce it to zero while the attribution efficiency is consistent with 100%.