Prince Yadav | Search arXiv

arXiv2026

The "Rubin 2036: Opportunities for Discovery" Workshop: A Community Roadmap for the Future of the Vera C. Rubin Observatory After the LSST

The NSF-DOE Vera C. Rubin Observatory will complete its ten-year Legacy Survey of Space and Time (LSST) in the mid-2030s. Its 8.4 m aperture and 9.6 square degree field of view give it an étendue unmatched by any facility planned for that era. The Rubin 2036: Opportunities for Discovery workshop (KIPAC/SLAC, 3-5 August 2026) brought together approximately 100 participants to discuss Rubin's future after LSST. This paper summarizes the workshop's structure and discussions across five working groups. The continued-surveying group explored a machine-learning foundation model pretrained on Rubin imaging for detection, deblending, and classification. The cadence group considered allocating post-LSST observing time and strategies for transiting exoplanets, Type Ia supernova cosmology, time-delay cosmography, and second-scale transients. The filter group explored broadband, medium-band, and narrowband filters for photometric redshifts, high-redshift large-scale structure, AGN reverberation mapping, Galactic archaeology, and emission-line imaging. The multi-messenger group explored follow-up of next-generation gravitational-wave and neutrino sources, including standard sirens, kilonovae, and Galactic supernovae. The technology group organized options by cost and lead time, including fast readout, new filters, a faster shutter, and summit edge computing. Cross-cutting topics included AI/ML, Chilean and Latin-American co-leadership, satellite megaconstellations, and weather and climate trends at Cerro Pachón. White papers and workshop talks are available through https://zenodo.org/communities/rubin2036workshop/ and https://rubin-observatory.github.io/Rubin-Beyond-LSST/pages/Rubin2036proceedings/.

astro-ph.IM↗

arXiv2026

DeepDISC on LSST Data Preview 1: Scene-Level Detection, Deblending, and Star/Galaxy Classification in the ECDFS and EDFS Fields

We present the first application of DeepDISC, a scene-level deep-learning framework, to real LSST data, using the Data Preview 1 (DP1) coadded images of the Extended Chandra Deep Field-South and the Euclid Deep Field-South fields. In a single pass, DeepDISC jointly detects, segments, and classifies sources as stars or galaxies, rather than treating these tasks as separate steps as in traditional pipelines. We warm-start from a model pretrained on LSST simulations and fine-tune it on DP1 six-band images, using the DP1 catalog's extendedness flag (point-like versus extended) as provisional class labels. The classification part of the network is then fine-tuned alone on a curated sample of stars and galaxies with spectroscopic, Gaia, and photometric identifications, so the final output is a star/galaxy classification. Fine-tuning on real DP1 observations substantially improves source recovery relative to the un-fine-tuned model, raising galaxy and star completeness from 40.2% and 13% to 84.2% and 49.1% under strict class-aware matching; under class-agnostic matching the fine-tuned model reaches 84.9% and 74.3%. On a common sample of sources that neither classifier saw in training, DeepDISC classifies to within about one percentage point of a published photometry-based Random Forest for the same field (97.4% vs. 98.5%); at fainter magnitudes DeepDISC labels more sources as stars than the Random Forest, and independent labels favor the Random Forest in most of these disagreements. Cross-matching to deeper HST/CANDELS imaging on a held-out footprint gives an unrecognized-blend fraction for DeepDISC of 14.9% (95% CI 12.5-17.8%), consistent with or modestly below the 18.0% we measure for the production DP1 catalog. This work establishes a baseline for DeepDISC on real LSST data and a foundation for self-supervised pretraining and for LSST Data Preview 2.

astro-ph.IM↗