Abstract

Utilizing the Cannon to Predict Stellar Parameters

Student: Derek Shank
Mentor: Timothy Beers (University of Notre Dame Department of Physics)

Stars have physical properties, such as temperature, which can be observed through data collected by large-scale astronomical surveys. The Cannon is a program which takes the collected stellar data and performs a series of tests to determine a star’s properties. In future work stellar properties can be determined at a rate far faster with The Cannon compared to individually analyzing each star.

The Cannon is a data driven analysis tool designed to predict stellar parameters based on a transfer of “stellar labels” from external datasets to large-scale spectral catalogues. The prediction of stellar parameters such as metallicity ([Fe/H]), effective temperature (Teff), carbon abundance ratios ([C/Fe]), and surface gravity (log g) will enable users to focus their attention on interesting stars which meet their criteria. Synthetic normalized spectra were used as the reference set to properly calibrate The Cannon for the test sets. In future work The Cannon will be applied to stellar spectra collected by the Sloan Digital Sky Survey (SDSS), the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST), and others. Comparison of the suggested stellar parameters from The Cannon with alternative approaches, such as the non-SEGUE Stellar Parameter Pipeline (n-SSPP) will greatly reduce the time required to obtain confident estimates.