About this Article

Karoline Gilbert (kgilbert[at]stsci.edu)

Published February 5, 2026

The Nancy Grace Roman Space Telescope remains on track to launch as early as fall 2026, which would lead to the start of its operations — and the arrival of the first Roman datasets — in early 2027. Multiple milestones on the road to launch have recently been achieved:

• Roman’s four community-defined surveys, three Core Community Surveys and the Galactic Plane General Astrophysics Survey, are fully specified.
• Thermal vacuum testing of the integrated instruments, telescope, and spacecraft assembly concluded in October 2025.
• The full observatory was successfully integrated into a single unit in November 2025 and is expected to ship to Kennedy Space Center as soon as June 2026.
• The first call for proposals for Roman data analysis funding, new Roman surveys, and more is now open, with proposals due March 17, 2026.

STScI recently launched the Roman Research Nexus, a cloud-based science platform developed to enable efficient access, exploration, and analysis of Roman’s large volume of data. You may access the Nexus now to explore Roman’s scientific capabilities and prepare to analyze Roman’s data as soon as operations start. Read on to learn more about the Nexus and its current functionality.

The Roman Research Nexus Supports Analysis of Roman’s Large Datasets

Roman’s Wide Field Instrument (WFI) has approximately 200 times the area of Hubble’s Wide Field Camera 3 (WFC3) IR channel. Roman will be capable of surveying the sky 1,000 times faster than Hubble, yielding in one month a data volume approximately twice that generated by Hubble over 30 years. All of Roman’s data will be non-proprietary and open access, generating a huge data volume stream for the science community.

The scale of Roman’s data will necessitate a different approach to data exploration, reduction, and analysis compared to the typical processes for Hubble and Webb data. While all Roman data will be archived and available in MAST using Amazon Web Services (AWS), Roman’s large data volume means that downloading and analyzing data on personal compute resources (e.g., laptops) will be prohibitive for all but the most limited analyses of Roman’s surveys. The Nexus resolves that by providing a rich computing environment that is co-located in AWS with the MAST datasets, eliminating the need for the egress of large datasets for most users.

Periodic, science-driven data releases will provide science-ready data products, including calibrated images, mosaics, spectra, light curves, ePSF libraries — and multi-band, spectroscopic, and variability catalogs. The data release products will reduce or remove individual data processing requirements for many users. These higher-level products will be archived in MAST and available through the Nexus, as will supporting data from other missions also hosted by MAST on AWS.

Providing a Rich and Flexible Scientific Computing Environment

The Roman Research Nexus is designed to meet the needs of scientists who will explore and analyze Roman’s data and data products, regardless of their expertise or experience. The Nexus is more than an online interface to Roman data: It is a computing platform that provides users with the necessary work environment to maximize their science.

Currently, the Nexus provides easy access to Roman software, simulation, visualization, and analysis tools; training tutorials; and simulated and ground-test data. The Nexus includes Roman-specific software and tools, such as RomanCal (the Roman image calibration and processing pipeline) and Pandeia (the python engine of the Roman WFI Exposure Time Calculator), and commonly used python libraries.

The Nexus also allows you to install your own software to bring your own tools to Roman’s datasets.

Prior to launch, simulated Roman WFI datasets and ground-test data are available in the Nexus. The Nexus provides access to the Open Universe simulation (Open Universe et al., 2025), which simulates spatially overlapping imaging surveys that will be carried out by Roman and the Vera C. Rubin Observatory. It also hosts small Roman I-Sim datasets that include simulated detector-level calibration features, simulated grism images, and bright star ground-test images from thermal vacuum tests of the WFI.

The Nexus will also soon host simulated datasets for each of Roman’s Core Community Surveys. The Core Community Survey simulations are being produced using Roman I-Sim, and constructed using the detailed implementation for each survey as well as realistic astrophysical source catalogs as inputs.

What an Example Session on the Nexus Looks Like

After logging in to the Nexus, users will be provided with a pre-configured environment that already has pre-installed software and required Python libraries. If, for example, users are interested in quantitatively measuring properties of specific astrophysical objects in realistic simulated Roman images, they may utilize the provided WFI data simulation workflow. This workflow consists of a set of Jupyter notebooks that walk a user through the steps needed to simulate a Roman dataset with realistic astrophysical sources and instrumental effects included using Roman I-Sim, run the resulting data files through the Roman calibration and data processing pipeline, and analyze the resulting calibrated images.

The Nexus is built to be scalable. Users can spin up servers with the compute resources — CPU and memory — they need for their intended task, and can scale up when needed. (This happens within a user’s allocated resources. Team accounts, described below, enable a larger range of resources.)

For example, for sessions where users intend to simulate a sizeable Roman dataset, significantly more compute resources will be required compared to sessions where users intend to analyze an existing catalog derived from a simulated dataset. The Nexus provides persistent storage, which means files uploaded or created during sessions will remain in user directories across sessions.

Providing a Pedagogical Environment

The Nexus hosts a number of tutorials accessible in any session. Standalone Jupyter notebooks focus on specific tasks and include significant documentation. Sets of tutorial notebooks are organized in guided, end-to-end workflows. Currently workflows include WFI data simulation (described in the section immediately above), WFI data analysis, and WFI observation planning. A time-domain workflow is in development.

The workflow focused on WFI data analysis includes a series of Jupyter notebooks that steps users through the process of retrieving data held in MAST, working with ASDF files, and visualizing the data. The tutorial workflow then branches out, allowing users to learn how to perform various analysis tasks: performing aperture photometry, measuring galaxy shapes, making image cutouts, and extracting spectra from a wide-field slitless spectroscopy observation.

The workflow focused on WFI observation planning includes Jupyter notebooks that walk users through the process of estimating the signal-to-noise ratio for a large number of sources given various sets of observing parameters, understanding characteristics of the WFI (such as the point spread function and the filter response) and their impact on the data, as well as visualizing key elements of survey design.

Survey planning and exploration tools include a notebook that uses the output of an Astronomer’s Proposal Tool survey specification to plot detailed survey footprints, including maps of exposure time. A background visualization tool provides an estimate of the background level for a given target as a function of calendar day and optical element, which may inform special requirements and survey strategies for background-limited observations. The elements of the observation planning workflow may also be used to deeply explore Roman’s community-defined surveys and their suitability for pursuing various science cases.

Built For Collaboration

The Nexus is explicitly designed to enable scientific collaboration. Team accounts support groups that are working on a common project. They provide shared compute resources, software environments, and a team directory, which help coordinate development. Team accounts provide access to larger compute resources compared to individual user accounts.

Teams may also initiate a Real-Time Collaboration Server. These servers allow team members to work together on the Nexus, co-developing software, interactively debugging, and co-editing Jupyter notebooks simultaneously.

While everyone with a MyST account has immediate access to an individual account on the Nexus, a team account must be requested. Requests will be reviewed to ensure a direct connection to Roman science. Team accounts may be requested by any user, as described in RDox.

Available Now: Start Exploring!

The Nexus is currently in an early access phase, which was opened to the science community in mid-December 2025. While core Nexus functionalities are available, its features and available content will continue to evolve and grow. Feedback from early users will inform future development.

Future Roman users are encouraged to jump onto the Nexus now to support preparation of Cycle 1 proposals and become familiar with analyzing simulated Roman datasets in the cloud. The simulated Roman datasets, and existing tools and software provide an opportunity for the science community to explore tutorial-style workflows, understand Roman’s data structures, analyze simulated and test data, and build and test tailored workflows in anticipation of the beginning of science operations in 2027. All that is required to start working in the Nexus is a myST account.

Learn more about the Nexus, its key features, individual and team accounts, computing resource credits, and more on RDox.

The Roman Research Nexus is developed by STScI in collaboration with Mission Partners.