A new framework called CIGaRS, from the Institute of Cosmos Sciences at the University of Barcelona and published in Nature Astronomy, models multiple factors affecting Type Ia supernova observations to improve cosmological inferences without relying solely on spectroscopy.

CIGaRS aims to maximize data utility from next-generation surveys to advance understanding of universal expansion, dark energy, and supernova physics.

A key finding is that galaxy redshifts can be estimated accurately using imaging data alone, with precision approaching spectroscopy, addressing a major limitation for Rubin Observatory surveys where most objects are photometric.

The framework enables redshift estimation from imaging data to handle data growth from upcoming surveys, especially the Vera C. Rubin Observatory.

Lead author and co-author emphasize using Bayesian inference and AI to identify potential systematic biases and maximize information from photometric data.

The approach could boost cosmological constraint precision by up to a factor of four compared with traditional methods that rely on smaller spectroscopic samples.

Combining physics-based simulations with AI could improve cosmological constraints by up to fourfold relative to spectroscopic-limited methods.

CIGaRS uses simulation-based inference and Bayesian methods to simulate universes and train a neural network to map observations to underlying physical parameters, enabling analysis of tens of thousands of supernovae.

Simulation-based inference and neural networks train on simulated universes, allowing thousands of supernovae to be analyzed collectively and reducing reliance on spectroscopy.

CIGaRS integrates supernova physics, host galaxy properties, dust effects, historical supernova rates, and cosmic expansion in a single cohesive model to capture interdependencies often treated separately.

Type Ia supernovae are standard candles whose brightness is influenced by host galaxy properties; CIGaRS models these effects within a unified Bayesian, physics-informed framework.

The model connects supernova physics, host galaxy effects, dust, supernova rate evolution, and cosmic expansion in one integrated approach.