Using Software to Share Accurate Models of Galactic, Extra-Galactic Emissions with the Scientific Community

Published May 9, 2024

The figure shows pictures of some of the Astrophysical models available in PySM, each image is a full-sky map simulated with PySM of the intensity of each emission (color-coded based on the colorbar below each image) at a specific electromagnetic frequency. View the code used to generate the images. Credit: Andrea Zonca

By SDSC Communications

Last year, a team of U.S.-based scientists were awarded a $52.7 million research grant from the U.S. National Science Foundation (NSF) to expand the Simons Observatory. Located in the Atacama Desert of Chile, the remote high-altitude astronomy observatory investigates the origins of the universe by measuring the first electromagnetic radiation ever emitted after the Big Bang. With the additional NSF support, the Advanced Simons Observatory under this new name provides more detectors, a solar energy system, funding for scientists developing the data analysis pipeline, and training of early-career researchers and students through mentorship opportunities with senior researchers at the observatory.

The San Diego Supercomputer Center at UC San Diego contributes to this project through the efforts of Andrea Zonca, lead of the Scientific Computing Applications group at SDSC. Zonca oversees the development of a software package known as Python Sky Model, or PySM. It serves as a catalyst for the team to implement accurate models of galactic and extra-galactic emission and share them with the rest of the scientific community.

“PySM uses the Python programming language to implement community-developed models based on real data. The package is capable of simulating the emissions in the microwave regime of our own galaxy, for example photons emitted by electrons as they propagate in the galactic magnetic field, or from extra-galactic sources, for example far away galaxies,” Zonca said. “Instead of having each experiment in the field of cosmology implement their own simulations, the purpose of this package is to take advantage of the expertise of scientists across all current experiments and develop the best possible astrophysical emission models to be released for free to the community to be used in simulations of any number of experiments in the field.”

Zonca said that the first version of PySM, released under an open source license by the University of Oxford (UK) in 2016, has become a community-developed software openly developed on Github with extensive documentation that tracks all details about how the models have been developed to ensure reproducibility.

For the observatory, the University of Pennsylvania builds the hardware upgrades, while a team of experts from UC Berkeley, the Flatiron Institute in New York and Princeton University perform the data analysis.

Zonca (Ph.D., University of Milano) has been working in the field of cosmology for 17 years, including as a postdoctoral scholar at UC Santa Barbara from 2009 to 2013, when he joined SDSC. In 2015, he was funded by UC Berkeley to work in data analysis of cosmic microwave background experiments and was named as a member of Berkeley's Computational Cosmology Center.

“Thanks to the additional NSF funding, the PySM software will continue to be improved, and it will be used to create realistic simulations of what the actual experiment will measure, in order to develop and test our data analysis pipelines,” Zonca said. “Even when the telescopes deliver real data, we will continuously compare data to our models and use simulations to assess the reliability of the Advanced Simons Observatory results.”

This work has been funded by the NSF (award no. 2153201).