Ryan Camilleri and Tamara Davis tell us about how they have examined models beyond ΛCDM using the Dark Energy Survey’s wonderful supernova catalogue. Tantalisingly, they find that a number of models are “moderately preferred” over ΛCDM (in model comparison speak).
They also, very admirably, check whether crucial aspects of the DES pipeline are model dependent or not. They find that, so long as the reference model is close-ish to the true model then the pipeline is accurate. “Close-ish” is very generous here as well, as they even found in simulations that when one processed the data with models 10σ from the truth, the subsequent parameter constraints were still within 1σ of the truth. The moral of this is that, even though the supernovae were processed assuming ΛCDM, so long as the true cosmology isn’t too far from ΛCDM then this doesn’t matter.
This means, if you have your own model that they haven’t tested, you don’t need to simulate the entire DES analysis pipeline to analysis your model, you can do your model comparison at the level of the Hubble diagram. Nice!
Paper: arXiv: 2406.05048
Cosmology Talks 