Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

OpenUniverse2024 Workshop

Hands-On Science with OpenUniverse2024 Roman and Rubin Simulations

OpenUniverse2024 Introduction

OpenUniverse2024 is a collection of large-scale simulated observations designed to resemble the expected data products of the Nancy Grace Roman Space Telescope and the Vera C. Rubin Observatory. Both observatories are simulated over the same roughly 70 square degree patch of sky, so the same objects appear in Roman’s high-resolution near-infrared imaging and in Rubin’s wide optical ugrizy imaging, and the two can be analyzed together. The dataset includes wide-area imaging, a time-domain survey that revisits a smaller footprint many times, and mock source catalogs. For Roman there are truth catalogs of the input object properties, noiseless truth images, calibrated single-epoch images, and coadds. For Rubin there are raw pixel data, calibrated exposures, coadds, and photometric catalogs.

All data are public, hosted on AWS in the S3 bucket s3://nasa-irsa-simulations/openuniverse2024/ (us-east-1 region), and require no credentials. Catalogs are stored in Parquet, hdf5, and fits format. Images are multi-extension FITS.

If you use the OpenUniverse2024 dataset in your work, please follow the “How to Cite” instructions on the Registry of Open Data on AWS: OpenUniverse2024 page.

Workshop Resources

The below materials were presented in a September 2026 workshop that provided a hands-on introduction to working with the OpenUniverse2024 dataset. Rather than surveying the simulations from a distance, we worked through Python Jupyter notebooks that access the data directly in the cloud and carry it through to a scientific result — locating a transient, building a light curve, fitting a spectral energy distribution. The emphasis throughout is on combining complementary survey data: what you can measure by using Roman and Rubin together that you cannot get from either one alone, and how to structure that work as a scalable cloud workflow rather than a local download.

Agenda

The workshop consisted of the following five segments. Slides and videos are provided where available.

What Is OpenUniverse2024?

Slides (PDF)

Recording of the talk “What Is OpenUniverse2024?” by Jessica Krick.

Introduction to Fornax

Slides (PDF)

Recording of the talk “Introduction to Fornax” by Troy Raen.

Accessing OpenUniverse2024 Data

Notebook walk-through (no slides).

Recording of the “Quickstart” notebook walk-through by Jessica Krick.

Visualizing OpenUniverse2024 Data

Notebook walk-through (no slides).

Recording of the “Firefly Visualization” notebook walk-through by Jaladh Singhal.

Self-paced tutorial exploration

Participants worked through notebook tutorials at their own pace (no slides or video).

Before You Start

The workshop runs entirely in the Fornax Science Console, NASA’s cloud-based science platform, so there is nothing to install on your own machine. The Fornax Quick Start Guide walks through the five steps of getting an account, logging in, starting a server session, opening a notebook, and shutting the server down again.

Tutorials

Workshop participants will explore the following tutorials, all part of the IRSA’s Simulated Data Tutorials. They assume some familiarity with Python and Jupyter notebooks, but not with Roman, Rubin, or cloud data access.

The tutorials are available on the Fornax Science Console in the directory ~/fornax-notebooks/irsa-tutorials/simulated-data/OpenUniverse2024/. They should automatically connect to the irsa-tutorials environment kernel, which has all libraries and other dependencies pre-installed.

To run the tutorials outside of Fornax, first clone the irsa-tutorials repo. Then, follow the instructions in the first code cell of each notebook to install the dependencies.

Quickstart

Access OpenUniverse2024 Roman and Rubin images and catalogs.

Firefly Visualization

Use Firefly to explore the structure of the simulations and visualize their content.

Roman Time Domain Survey Supernova

Explore the Roman Time Domain Survey observation sequence and animate a simulated SN Ia across epochs.

TDE Light Curve

Locate a simulated TDE, retrieve Roman images, and build a multi-epoch light curve.

SED Fitting

Fit spectral energy distributions for supernova host galaxies using Prospector.

Questions

Get help by posting in the Fornax Helpdesk (public) or the OpenUniverse Workshop category (private to workshop participants and Fornax staff).