A formal update of the ExoMol Database Collection

Abstract The ExoMol project provides spectroscopic and other key data for the analysis of hot astronomical objects with a particular emphasis on studies of exoplanet atmospheres. ExoMol has developed its own data structure designed for efficient delivery of what can be very large datasets. This data structure has evolved as data requirements have changed with new astronomical techniques, such as the requirement of high resolution cross correlation spectroscopy, the need to treat hyperfine structure, and developments such as the widespread use of machine learning. In response to these issues, we present a formal update to the data formats used to present ExoMol data. This includes a move towards internationally adopted data standards where they exist and the use of JavaScript Object Notation (JSON) files to present metadata for the datasets. At the same time, thorough cleaning and re-evaluation of metadata has been undertaken. A summary of the various databases provided by the ExoMol Database Collection is presented and their data structures are discussed. The prospects for future data provision are considered.