Abstract:

The Harvard-Smithsonian Center for Astrophysics manages the comprehensive paper database Astrophysics Data System (ADS), which hosts nearly 15 million records, each with detailed citation and impact metrics. The Astronomy Image Explorer, hosted by AAS, is a database of images and figures published in peer-reviewed astronomy journals. In these records, the use of color in figures began in the mid-1990s and has become generally conventional since then. Using glue, a Python library that explores relationships within and between related datasets, we generated color distribution histograms of figures in AIE for ADS-listed journal articles in color figure-heavy fields of astronomy, such as stellar formation and galactic evolution. We correlated the color distributions with impact metrics from ADS. We predict results that certain RGB distributions improve article comprehensibility up to a certain threshold.