Abstracts
Illuminating Hidden Binary Parameter Space with Roman Microlensing
By: Natasha Abrams
Type: Talk
Abstract: Gravitational microlensing provides a unique opportunity to probe binary stars, black holes, neutron stars, white dwarfs, and brown dwarfs. Microlensing is sensitive to non-luminous objects, extreme mass ratios, and intermediate semi-major axes, as it has a complementary selection function to other binary detection methods. Unlike other surveys before it, Roman will measure the photometric and astrometric microlensing signals jointly. This will enable us to distinguish more binary-lens microlensing events from single-lens events and more robustly constrain the binary’s masses and semi-major axes. We performed a mock-Roman microlensing simulation using PopSyCLE. We will present how the measured binary fractions from Roman microlensing will differ from ground-based surveys, how extinction impacts the measured binary fraction, and compare the intrinsic binary parameter distributions to the anticipated measured distributions from Roman microlensing events.
Co-Authors: Jessica Lu, Sage Remulla, T. Dex Bhadra, Macy Huston
Uncovering Starless Dark Matter Halos in the Era of Roman
By: Gagandeep S Anand
Type: Talk
Abstract: A cornerstone prediction of the ΛCDM model is the existence of gas-filled starless dark matter halos on sub-galactic mass scales. FAST observations previously uncovered a compact HI Cloud (hereafter Cloud-9) in the vicinity of the nearby spiral galaxy M94. Independent observations by both the VLA and GBT confirmed the cloud and the refined its properties. Cloud-9 is compact (1.4 kpc), dynamically cold (HI line width of 12 km/s), non-rotating, and fairly massive (HI mass of 10^6 M☉). Deep optical observations with the Hubble Space Telescope recently ruled out the presence of a dwarf galaxy down to 10^4 M☉ with 99.5% confidence. The non-detection of a stellar counterpart reinforces the interpretation that Cloud-9 is a starless dark matter halo filled with hydrostatic gas in thermal equilibrium with the cosmic ultraviolet background. Future studies of a larger population of such starless halos will allow for the determination of fundamental dark matter halo properties, providing valuable insight into the nature of dark matter itself. However, these objects are expected to be rare gems, and finding more presents a notable challenge. One of the many benefits of a large Roman WFI Survey of nearby galaxies with significant areal coverage would be to uncover more of these starless halos in coordination with blind HI surveys.
Co-Authors: Alejandro Benitez-Llambay (University of Milano-Bicocca), Rachael Beaton (STScI), Andrew Fox (STScI), Julio Navarro (University of Victoria), Elena D'Onghia (University of Wisconsin, Madison)
Toward Wide Field Slitless Spectroscopy with Roman: Processing WFC3/IR Grism Data in Moderately Croweded Fields
By: Rachael L Beaton
Type: Poster
Abstract: Wide-field slitless spectroscopy with Roman has great potential as a tool to characterize stars across the Milky Way galaxy and its nearby satellites. However, historical use of this observing technique has been largely limited to high-latitude fields focusing on galaxy applications with some work extracting serendipitous observations of very cool stars. In this contribution, we describe our efforts extracting slitless spectroscopy across 60 unique WFC3/IR Grism observations (with both grisms) taken in a "snapshot" mode. For each field, we extract spectra using different filter/grism combinations, stitch together both grism results, and characterize the degree of contamination from neighboring sources. Using "cleanest" spectra in our coverage we use the Canon to apply label transfer between Gaia measurements and our spectra. We demonstrate the ability to produce basic stellar characterization from the Grism spectra (temperature, surface gravity, metallicity). We articulate challenges to expanding this work to the areas and spectral diversity for Roman as well as identifying improvements for our adopted techniques.
Co-Authors: Rachael Beaton (STScI), Maya Seagraves (Cal Poly San Luis Obispo), Sten Hasselquist (STScI), Meredith Durbin (UC Berkeley/Univ of Washington)
A Deep, Wide Search for Solar System Objects
By: Susan Benecchi
Type: Talk
Abstract: We present the scientific return from a deep (r~30.5) and wide-field (1.12°) survey using NASA’s upcoming Roman Space Telescope along the trajectory of NASA’s New Horizons (NH) spacecraft. Although the probability of discovering the next Kuiper Belt Object (KBO) for a close NH flyby is ~3.5%, the ensuing encounter would yield a unique, scientific bonanza of extreme value that cannot otherwise be obtained for many decades. Such a discovery would enable groundbreaking comparisons between the geology, geophysics, and surface composition of KBO Arrokoth, visited by NH in 2019, to those of a KBO located approximately twice as far from the Sun. This survey would discover ~16 KBOs observable from NH’s trajectory at a range of large phase angles to determine shapes, surface scattering and microphysical properties, and to search for satellites at resolutions neither HST nor JWST can achieve. This same survey also advances understanding of the KBO size frequency distribution (SFD) at the small size end, a phase space inaccessible to ground-based surveys, by discovering >5000 new, small KBOs with an order of magnitude increase in the number of objects in the critical H=11-14 range (d≤20 km) from which the SFD is evaluated. It will also be sensitive to a multitude of near earth and main belt asteroids. Individually and cumulatively, these studies have significant implications for solar system formation models and cratering history. A more complete understanding of our local small body population by extension may likewise yield insight for exo-planetary disk studies.
Co-Authors: Susan Benecchi, Pontus Brandt, Richard Cosentino, Wesley Fraser, Will Grundy, Bryan Holler, Takashi Ito, JJ Kavelaars, Casey Lisse, Lowell Peltier, Rosemary Pike, Simon Porter, Kelsi Singer, S. Alan Stern, Anne Verbiscer, Fumi Yoshida
The Roman Galactic Exoplanet Survey: Cool Exoplanets Dominate
By: David Bennett
Type: Talk
Abstract: The Nancy Grace Roman Space Telescope was selected by the Astro2010 Decadal Survey "to settle essential questions in both exoplanet and dark energy research." These essential exoplanet questions are to be settled with the Roman Galactic Exoplanet Survey, which will complete the statistical census of exoplanet systems with a gravitational microlensing study of exoplanets in Earth-like and wider orbits to complement the Kepler mission's comprehensive study of planets in short period orbits. Since the selection of Roman by the Astro2010 Decadal Survey, multiple ground-based microlensing surveys have confirmed earlier suggestions that super-Earths and Neptunes are much more common in Jupiter-like orbits and found that there is no prominent sub-Saturn mass-ratio "exoplanet desert", which was predicted by some versions of the core accretion theory. The most surprising result has been the 2023 discovery that low-mass free-floating or very wide orbit planets appear to be very common. The first estimate of the mass function of these candidate free-floating planets indicates that they are approximately 6 times more common than all other known categories of exoplanets combined. This analysis suggests that the RGES program should detect about 1000 of these free-floating or very wide orbit planets, and that these detections are likely to be dominated by sub-Earth-mass planets. Very recent observations and also demonstrated that Roman discoveries combined with simultaneous observations from other telescopes can determine the masses of these free-floating planet candidates.
Co-Authors: The Roman Galactic Exoplanet Survey Project Infrastructure Team
Absorption Line Study of JWST Galaxies
By: Anukalpa Bhaumik
Type: Poster
Abstract: Galaxies continuously evolve over cosmic time through interactions with the surrounding circumgalactic media (CGM). Here, we present an absorption-line study of the CGM associated with three high-redshift galaxies observed by the James Webb Space Telescope. These galaxies with emission redshift in the range 1.9-3 were observed in the GOODS-South field as part of the JWST Advanced Deep Extragalactic Survey (JADES). Analysis of absorption spectra available under Data Release 3 provides insight into the kinematics, depletion patterns, and gas conditions in these galaxy environments, with a focus on key metal lines such as Ca II, Mg II, and Na I. The kinematic analysis reveals significant velocity widths in the galaxies, ranging from 400 to 2000 km/s, with absorption features indicating large-scale gas motions possibly linked to strong outflows, turbulent CGM dynamics, or AGN-driven winds. Elemental abundance and depletion patterns in these environments appear to be quite different from the local interstellar medium. These results, which are part of an ongoing study, shed more light on the role of galactic feedback, AGN activity, and dust processing in shaping the CGM at high redshifts. Probing more such sightlines could enable a richer understanding of CGM-galaxy interactions across the history of the Universe.
Co-Authors: Dr. Katherine Rawlins
From Dozens to Thousands: Stellar Streams with the Roman Space Telescope
By: Ana Bonaca
Type: Talk
Abstract: Tidally dissolving satellite dwarf galaxies and globular clusters produce long, thin and kinematically cold stellar streams, which are direct tracers of hierarchical galaxy growth, as well as dynamical tracers of dark matter in galactic halos. Thanks to wide-field imaging surveys, dozens of such streams are now known in the Milky Way and in nearby galaxies. However, the current depth and resolution only allow detection of the most prominent and the most nearby streams, thus limiting them as probes of galaxy formation physics and cosmology. Roman provides unique capabilities for increasing photometric depth and angular resolution simultaneously over wide areas of the sky. I will discuss how Roman's planned surveys, together with future GO programs, can unveil a full census of streams in the Milky Way and the Local Universe, recover the shapes of individual dark matter halos, and reveal dark matter substructure.
The SCExAO/OASIS Survey: Imaging New Exoplanets and Providing Targets for the Roman Coronagraph Technology Demonstration
By: THAYNE M CURRIE
Type: Poster
Abstract: We describe the current status of the Observing Accelerators with SCExAO Imaging Survey (OASIS), the largest active exoplanet direct imaging survey, supported by NASA HQ to identify targets for the Roman Coronagraph technology demonstration. OASIS uses SCExAO/CHARIS high-contrast integral field spectroscopy and Keck/NIRC2 thermal infrared imaging to target 150
young stars whose Hipparcos and Gaia astrometry reveal dynamical evidence for an unseen substellar companion. The first OASIS discoveries include multiple new, confirmed discoveries of exoplanets (e.g. HIP 54515 b) and brown dwarfs (e.g. HIP 71618 B): all with simultaneous atmospheric, orbital, and dynamical mass constraints absent from traditional surveys.
Finally, we describe the critical role this program plays in supporting the Roman Space Telescope Coronagraphic Instrument. OASIS has now provided the first demonstrably suitable system for Roman Coronagraph technology demonstration, HIP 71618, which has now been selected for observations with the HLC. It and HIP 54515 b have been selected for spectroscopy observations technology demonstration observations. We describe new OASIS observations of HIP 71618/HIP 54515 relevant for the Roman Coronagraph and results from a complementary program to vet PSF reference stars that can be paired with these two companions.
Co-Authors: Mona El Morsy
Effortless: Efficient Optimal Image Reconstruction for Roman and Beyond
By: Kaili Cao
Type: Poster
Abstract: The forthcoming Nancy Grace Roman Space Telescope will revolutionize astrophysics by generating huge amounts of data of unprecedented quality. To properly address the data deluge and fully realize its potential, analysis tools that are both efficient and optimal are needed. For weak gravitational lensing cosmology, a major challenge is that native images are undersampled and need to be reconstructed to enable accurate measurements. Effortless (EFFicient Optimal image ReconsTruction using LESS memory; previously known as Fast IMCOM) is a new algorithm for linear image reconstruction. Like its predecessor IMCOM, it offers control over point spread functions in output images; by avoiding laborious calculations, it is tens of times faster and has a smaller memory footprint.
In this talk, I will present promising first results of apply Effortless to simulated Roman images. With oversampled native points known a priori, our ability to reconstruct undersampled images is not limited by the Nyquist–Shannon sampling theorem. A single image reconstructed by Effortless can lead to better measurements than a set of ~6 images coadded by IMCOM. While both algorithms were originally designed for weak gravitational lensing cosmology, Effortless can benefit studies of static features and dynamic aspects of the Universe alike. Moreover, the efficiency and interpretability of Effortless provides new possibilities for further reducing errors in measurements.
Co-Authors: Roman HLIS Cosmology PIT
Fisher Forecasts for Cosmological Yields from 3×2pt Analysis of the Roman High Latitude Imaging Survey
By: Kaili Cao
Type: Poster
Abstract: NASA's Nancy Grace Roman Space Telescope is currently scheduled for launch in August 2026. Its High Latitude Imaging Survey (HLIS) will provide powerful tests of cosmological models through sensitive measurements of cosmic shear, galaxy-galaxy lensing (GGL), and galaxy clustering. In this poster, I will present our Fisher forecasts of cosmological parameter constraints from combinations of the three probes mentioned above, carried out as part of the HLIS PIT's Data Challenge 1 (DC1). We find good agreement between Fisher analysis and Markov chain Monte Carlo (MCMC) analysis of the DC1 baseline data vector, and we exploit the flexibility of Fisher analysis to investigate varied priors on cosmological parameters and breakdown of contributions from different tomographic bins and different angular scales.
Given the benchmark DC1 priors, the forecast constraints from GGL+clustering are substantially stronger than those from cosmic shear, with the combination of all three probes ("3×2pt") providing moderate further improvement. Adding tight external priors on the power spectrum shape parameters ns, Ωb, and h0 can improve the (σ8, Ωm) FoMs by factors of 1.2–3.5. The smallest scale angular bins provide much more information than the largest scale bins, and the highest redshift tomographic bins provide more information than the lowest redshift bins. In the associated paper, we also explore varied priors on nuisance parameters describing unknown biases in photometric redshifts or shear measurements.
Co-Authors: David H. Weinberg, Vivian Miranda, Nihar Dalal, Tim Eifler, Jiachuan Xu, Haley Bowden
SUPERBACK: A WFI Science Program For Background Modeling and Subtraction
By: Timothy Carleton
Type: Talk
Abstract: The vast majority of photons seen by Roman's WFI will not be from distant galaxies, but large-scale, diffuse foregrounds from within our Solar System and the Milky Way. The precise extragalactic and cosmology science conducted with Roman depends on an accurate subtraction of these foregrounds. Unlike small FOV telescopes like HST and JWST, the background of Roman WFI images is expected to have complex structure and vary significantly across the field. I will present an overview of SUPERBACK, a WFI program to model the large-scale foregrounds in WFI images. Rather than empirically estimating these backgrounds, SUPERBACK takes an approach of forward-modeling each astrophysical component - Zodiacal Light, Diffuse Galactic Light, Wide Angle PSFs from bright stars, and thermal emission from the telescope. This approach promises to result in flatter backgrounds than empirical maps of the background and improve photometry of extended galaxies. In addition, this approach will retain the details of these foregrounds, enabling astronomers to analyze them (improving models of Zodiacal Light, for example).
Co-Authors: Seth Cohen, Qing Liu, Rosalia O’Brien, Rogier Windhorst, Russell Ryan, Alex Pigarelli, Rafael Ortiz III, Rolf Jansen, Brenda Frye
Resolving the supernova UV crisis with Roman: Enabling science from z > 4 supernovae
By: Jeff Cooke
Type: Talk
Abstract: Core-collapse supernovae (CCSNe) and superluminous supernovae (SLSNe) are UV-luminous, have been detected to z ~ 5 and can be detected to z ~ 20. They act as one of our only probes of the early Universe. They are brighter than their host galaxies, and they have high utility in galaxy evolution (outflows, inducing star formation, contribution to reionization, etc.), stellar evolution (massive star formation, explosion mechanisms, high-energy physics, etc.), and absorption-line research (host galaxy ISM, CGM, and line-of-sight IGM), have promise as standardisable candles, and trace the cosmic star formation, IMF, and chemical enrichment. Over 100 z > 2 CCSNe and SLSNe have been detected in existing deep images, as well as several lensed events, and Roman is expected to detect 1000s of z > 4 events. Yet we are not (and will not) be able to do science with them. This is because all supernovae are classified by their rest-frame optical light and events at z > 4 are detected only by their rest-frame UV light. At z > 4, identifying rest-frame optical features are redshifted out of the near-IR, and out of the mid-IR for the highest redshifts. In order to enable science with all existing and future z > 2 CCSNe and SLSNe, we need a supernova UV classification scheme. Roman, combined with deep, wide-field ground-based optical surveys, can resolve this problem by enabling the connection of rest-frame optical light curves, and spectroscopic properties via follow-up observations, with the rest-frame UV with curves and properties for events at z ~ 2-4. I will discuss how this can be done with Roman community surveys and small programs and I will explain how, non-intuitively, this work is easier and best done at z ~ 2-4 compared to low redshift.
Building a Galaxy-Star Classifier for Roman Space Telescope Alert Streams
By: Mario Damiano
Type: Poster
Abstract: NASA’s Nancy Grace Roman Space Telescope will open a new near-infrared time-domain discovery space, repeatedly imaging large sky areas to reveal supernovae, stellar variables, tidal disruption events, kilonovae, and other transient phenomena. To make these discoveries actionable, alert streams must be classified quickly enough to support filtering, prioritization, and follow-up. A fundamental early decision is whether an alert is associated with a galaxy-like or star-like source, since this distinction affects transient vetting, host association, variable-star rejection, and allocation of scarce follow-up resources.
We present a multimodal galaxy-star classifier developed for simulated Roman alert packets and associated truth-label products. The training set is derived from Roman image-differencing simulations, with each alert represented by science, template, and difference-image cutouts, together with tabular source-measurement features such as signal-to-noise ratio, morphology, fit quality, source extent, and magnitude. The model combines a convolutional neural network branch for image stamps with a parallel tabular-feature branch, trained with robust normalization, data augmentation, standardized metadata features, and class-weighted optimization.
On a held-out binary galaxy-star test set, the classifier achieved 99.49% accuracy, 0.9886 macro-F1, and 0.9991 ROC AUC, with strong performance for both galaxies and stars. These results demonstrate that compact multimodal classifiers can provide low-latency source-type scores for Roman alert filtering, enabling faster identification of high-value transient candidates and more efficient follow-up in the Roman era.
Co-Authors: Dr. Umaa Rebbapragada Dr. Ashis Mahabal
The Small Scale Structure of Dark Matter Through Roman’s Lenses
By: Tansu Daylan
Type: Talk
Abstract: One of the most stringent tests of Lambda Cold Dark Matter at sub-galactic scales is the census and characterization of substructure in lensing galaxies. With its wide field of view and high near-infrared sensitivity, the Nancy Grace Roman Space Telescope will significantly advance our ability to characterize the microphysics of dark matter via gravitational imaging by dramatically increasing the number of high-signal-to-noise strong lenses. Building on our previous work, we simulate dithered exposures using pixel-based HST-COSMOS galaxy sources and line-of-sight dark matter halos in the lensing model, producing a comprehensive suite of synthetic Roman images of strong lenses that reveals modeling degeneracies and identifying survey-design choices to help mitigate them. In addition, we report on our Roman Strong Lens Data Challenge, which has started receiving community submissions. Following its launch and commissioning, Roman observations of strong lenses at high galactic latitude will enable a sensitive census of substructure around extended arcs in hundreds of systems, yielding statistical sensitivity to the low-mass end of the dark matter subhalo mass, providing a statistical probe and characterization of subhalo concentrations.
Co-Authors: Simon Birrer, Bryce Wedig, Aysu Ece Sarıcaoğlu, Alan Huang, Francis-Yan Cyr-Racine, Cora Dvorkin, Douglas P. Finkbeiner, Alan Huang, Xiaosheng Huang, Rahul Karthik, Narayan Khadka, Priyamvada Natarajan, Anna M. Nierenberg, Annika H. G. Peter, Justin D. R. Pierel, Xianzhe TZ Tang, Jodie Xiao, Risa H. Wechsler
Galaxy Environments and Large-Scale Structure with Roman: A Machine Learning Approach
By: Jiani Ding
Type: Poster
Abstract: Understanding how galaxy evolution depends on environment remains one of the central open questions in astrophysics. While local processes tied to halo mass and galaxy density are well established, the cosmic web of filaments, clusters, and voids may also directly regulate star formation, gas accretion, and morphological transformation. Testing this across cosmic time has been difficult because existing surveys lack the depth and spectroscopic completeness needed to simultaneously reconstruct large-scale structure and characterize galaxy populations across cosmological volumes.
I will show that a deep learning approach trained on JWST slitless spectroscopy can efficiently identify emission-line galaxies with high precision while dramatically reducing the need for manual inspection, with learned representations capturing rich information relevant for a wide range of downstream science.
By applying this framework to Roman, I will highlight how large homogeneous emission-line galaxy catalogs from the High Latitude Wide Area Survey can be used to study diverse galaxy science, including reconstructing cosmic filaments, mapping overdense structures and protocluster candidates, and delivering a statistical test of whether cosmic-web connectivity directly regulates galaxy star formation, stellar mass assembly, and morphological evolution across cosmic time.
Co-Authors: Minghao Yue, Xiaohui Fan
Planning Reflected Light Observations of Known Exoplanets with the Roman Coronagraph
By: Clarissa R. Do O
Type: Poster
Abstract: The Nancy Grace Roman Space Telescope Coronagraph Instrument (CGI) will demonstrate key technologies required for reflected light imaging of mature exoplanets, making careful target prioritization and observation planning essential before observations are scheduled. This work presents a Roman CGI target planning pipeline that translates existing radial velocity and astrometric orbit constraints into the quantities needed to evaluate when known planets are most observable in reflected light. Starting from updated orbit fits, the pipeline samples the allowed orbital solutions for each target and propagates them across an observing window. For every posterior sample and epoch, it computes the projected separation, orbital geometry, phase angle, orbital radius, inferred planet properties, and posterior weights needed to predict reflected light observability. These posterior point clouds are then converted into photometric observables using either a simple prescription with geometric albedos and Lambertian phase functions or a treatment informed by atmospheric models, allowing detectability estimates to incorporate realistic dependencies on planetary properties. By comparing each sample against Roman CGI contrast curves as a function of separation, the pipeline estimates the time dependent probability that a planet will be detectable in a given observing band. For epochs with non-negligible detectability, it interfaces with EXOSIMS/corgietc Roman CGI noise models to estimate the integration time required to reach a specified signal-to-noise ratio under optimistic and conservative CGI performance assumptions. The pipeline also evaluates Roman pointing constraints such as the solar keepout and Galactic bulge avoidance, so that visibility windows can be interpreted alongside orbital and photometric favorability. Finally, the pipeline summarizes the full posterior ensemble into weighted percentile tables and diagnostic plots showing orbit tracks, working angles, detection probability, separation, phase angle, flux contrast, and integration-time credible intervals. Together, these products provide a reproducible framework for identifying favorable Roman CGI observing epochs and prioritizing known exoplanet targets.
Co-Authors: Ellis Bogat, Óscar Carrión-González, Sarah Blunt, Vanessa Bailey, Marie Ygouf, Alexander Venner, Amanda Chavez, Arthur Vigan, Bertrand Mennesson, Beth Biller, Brianna Lacy, Bruce Macintosh, Chen Xie, Christine Chen, Dan Sirbu, Dmitry Savransky, Elodie Choquet, Gael Chauvin, Izzy Huckabee, Jason Wang, Jessica Gersh-Range, Jingwen Zhang, Johan Mazoyer, Jorge Llop-Sayson, Julien Girard, Justin Hom, Karl Stapelfeldt, Laurent Pueyo, Macarena Vega-Pallauta, Mark Marley, Masayuki Kuzuhara, Max Millar-Blanchaer, Nick Schragal, Ramya Anche, Samantha Hasler, Schuyler Wolff, Sophie Noiret, Toshiyuki Mizuki, Tyler Robinson, William Balmer, Wolfgang Brandner, and Zarah Brown.
Technology and Science Advancing Observations with Roman Coronagraph Informed by Ground-Based High-Contrast Imaging
By: Mona El Morsy
Type: Poster
Abstract: Ground-based ExAO surveys such as GPIES and SHINE have detected ~20 planets by direct imaging, but blind discovery rates remain low. Direct Imaging alone poorly constrains orbits and cannot measure masses, limiting insights into atmospheric evolution. The OASIS survey targets planets around accelerating stars using Subaru/SCExAO and CHARIS IFU. Building on HIP 99770 b—the first planet jointly discovered via astrometry and imaging. We present OASIS detection of HIP 54515 b, the first OASIS planet and third via combined imaging and astrometry, and HIP 71618 B, the only substellar companion to meet Roman CGI’s TTR5 and suitable for coronagraph tech demonstration. Future OASIS and other programs are expected to identify additional CGI targets, expanding TTR5 candidates and enhancing the mission’s scientific and technical return.
Co-Authors: Mona El Morsy, Thayne Currie, Brianna Lacy, Taylor L. Tobin, Qier An, Yiting Li, Ziying Gu, Masayuki Kuzuhara, Danielle Bovie, Dillon Peng, Jeffrey Chilcote, Olivier Guyon, Miles Lucas, Timothy D Brandt, Robert de Rosa, Tyler D Groff, Markus Janson, N. J. Kasdin, Julien Lozi, Christian Marois, Bertrand Mennesson, Naoshi Murakami, Eric Nielsen, Sabina Sagynbayeva, Nour Skaf, William Thompson, Motohide Tamura, Taichi Uyama, Alice Zurlo, Vincent Deo, and Sebastien Vievard
Identification of High-Redshift GRBs with the Roman Telescope
By: Hallie Mae Fausey
Type: Poster
Abstract: Gamma-ray bursts (GRBs) are powerful probes of the high-redshift Universe. Their extreme luminosities and relatively simple power-law spectra make them ideal for studying the chemical evolution of the Universe, early star formation, and the Epoch of Reionization.
To date, only a handful of high-redshift (z > 6) GRBs have been detected, and the true rate of high-redshift GRBs remains uncertain. During the era of Swift and Fermi, some high-redshift GRBs may have gone undetected due to lower sensitivities and more limited coverage in the softer X-ray regimes compared to newer missions. With the launch of Einstein Probe (EP) and the Space Variable Objects Monitor (SVOM), several new high-redshift GRBs have already been identified, suggesting that the rate of high-redshift GRB rate may be higher than previously observed. Additionally, some high-redshift GRBs may go unidentified due to a lack of follow-up, with only ~30% of Swift GRBs having redshift estimates. Obtaining a more complete set of redshift estimates will be essential for understanding the true rate of high-redshift GRBs.
The Nancy Grace Roman Space Telescope High-Latitude Time-Domain Survey (HLTDS) will monitor ~18 deg² (wide) and ~6.5 deg² (deep) fields with a 5-day cadence over the span two years. Near-infrared observations of GRBs within the HLTDS fields can provide photometric redshift estimates via identification of the Lyman-α (Lyα) dropoff, a steep loss of flux that marks the edge of the Lyɑ forest. We simulate GRB afterglow observations using the Roman HLTDS cadence, filter set, and sensitivity to assess redshift recovery capabilities for GRBs. We investigate the impact of brightness and fading on redshift constraints, evaluate the impact of contamination from low-redshift high-extinction GRBs. We also explore how shorter cadence and more photometric band coverage could improve redshift retrieval to aid in the design of future GRB surveys with the Roman telescope.
Co-Authors: Benjamin Rose
Roman WFI processing and data products
By: Henry Ferguson
Type: Talk
Abstract: This talk will outline the current plans for processing and releasing the WFI data products from the Science Operations Center. It will also discuss areas where further development is being considered, as a way to stimulate discussion for new ideas and for input on priorities.
Roman-Rubin Synergies for Ultracool Dwarfs
By: John Gizis
Type: Poster
Abstract:
We present an analysis of ultracool dwarf science in the Roman-Rubin wide field area survey. The Vera C. Rubin Observatory LSST survey will detect L and T dwarfs to hundreds of parsecs, far beyond existing wide-field near-IR and mid-IR surveys. The Roman High-Latitude Wide-Area Survey is only ~1/10 the area of LSST, but offers higher spatial resolution and near-IR coverage. We show the benefits for machine-learning training samples from the overlapping datasets. We show how this sample will allow tests evolutionary models, star formation history, and the initial mass function for very-low-mass stars and brown dwarfs.
Co-Authors: John Gizis and Easton Honaker
Weak lensing with Roman: from galaxy shapes to shear measurement
By: Axel Guinot
Type: Poster
Abstract: Understanding the nature of Dark Matter and Dark Energy is at the center of the Stage-IV surveys. Among cosmological probes, weak gravitational lensing has taken an important place and is a flagship measurement planned with the current surveys with the Euclid Space Telescope, the Vera C. Rubin Observatory and by the forthcoming Roman Space Telescope.
Weak lensing relies on the measurement of cosmic shear which is a faint but coherent distortion in the observed shape of galaxies induced by the distribution of (dark) matter along the line of sight. To date, these measurements were limited by statistical errors due to the finite depth and area of our surveys. We are now entering a new era where we will be dominated by systematic uncertainties. We hence need to improve our image processing methods to achieve our desired constraints on cosmology.
In this presentation I will describe our strategy to address these challenges using Roman observations. I will present implementations of current methods alongside new techniques aimed at improving computational speed and statistical power of the survey.
From JWST to Roman: A Physics-Informed AI Super-Resolution Framework for Grism Spectroscopy
By: Aryana Haghjoo
Type: Talk
Abstract: The Roman Space Telescope's High Latitude Wide Area Survey will deliver grism spectroscopy at R~460–900 for roughly ten million emission-line galaxies, but at this resolution key diagnostic features such as the Hα+[N II] complex remain unresolved across a wide redshift range, introducing systematic biases in redshift measurements and derived physical quantities. Assembling the survey-scale high-resolution spectroscopic data needed to overcome this limitation is observationally prohibitive. We introduce a three-stage physics-informed deep-learning framework for spectral super-resolution, trained on 1,187 paired JWST/NIRSpec prism–grating observations from JADES, that enhances low-resolution galaxy spectra by a factor of 10 in resolving power (R~100 to R~1000). The model infers a redshift from a coarse super-resolved intermediate, then applies a residual refinement stage that uses multi-head self-attention across emission-line tokens to learn inter-line relationships and predict physically interpretable line profiles. We then benchmark this pipeline against seven classical deconvolution methods, demonstrating a 30% reduction in global reconstruction error, superior line detectability and width recovery. Here we characterize the domain gap between JWST NIRSpec and Roman grism spectroscopy and present a transfer strategy. Applied to simulated Roman grism observations, the adapted framework reconstructs Hα+[N II] for millions of galaxies, enabling resolved strong-line diagnostics and reducing the dominant spectroscopic systematic on Roman's BAO scale.
Co-Authors: Shoubaneh Hemmati, Bahram Mobasher
TheFARRM: Forced-photometry for All Roman-Rubin Matches
By: Jesse Han
Type: Talk
Abstract: We introduce TheFARRM (Forced-photometry for All Roman-Rubin Matches), a pipeline that measures Rubin LSST fluxes at the positions of Roman sources to produce a value-added catalog of LSST photometry and colors for every Roman detection. Roman’s angular resolution enables robust star-galaxy separation and deblending, while LSST six-band photometry provides deep optical SED coverage for discovering resolved stellar systems and characterizing galaxies. TheFARRM is designed for rapid community delivery: it operates on Roman L4 catalogs and leverages S3DF resources where the LSST data are stored, enabling efficient processing as soon as the Roman catalogs become available. We additionally plan to apply the LSST DESC RAIL photometric-redshift framework to the resulting photometry, delivering fast-turnaround photo-z estimates suitable for general astrophysics. We envision these products being returned to the Roman SOC as a Level-5 value added catalog, while the Roman/Rubin leadership continue discussions regarding LSST data rights policies. We demonstrate the full pipeline with an end-to-end simulation in which an ultra-faint dwarf galaxy is injected into a realistic COSMOS-Web galaxy field, and rendered using the romanisim and LSST DESC imSim image simulators. We present the pipeline, its validation with realistic image simulations, and its implications for near-field cosmology and galaxy-evolution science with Roman. We emphasize that TheFARRM is optimized for rapid delivery of scientifically useful cross-survey data products for general astrophysics applications, rather than for precision cosmology analyses.
Co-Authors: Ani Chiti, Alex Broughton, Jim Chiang, Risa Wechsler, Phil Marshal
Modeling Galactic cosmic rays in Roman's Wide Field Instrument using the HAZARD simulator
By: Anthony Harbo Torres
Type: Poster
Abstract: At the Sun-Earth L2 Lagrange point, the Nancy Grace Roman Space Telescope will be exposed to the space environment without the protection of the Earth's magnetic field, with an expected rate of order 100 events/detector/second. Furthermore, the Wide Field Instrument on-board Roman will sport H4RG-10 detectors with smaller physical pixels than the James Webb Telescope's H2RGs (10 vs 18 µm), so it will resolve more of the track structure. To test the effectiveness and residual statistical effects of cosmic ray rejection, we are developing a model of cosmic rays that goes beyond general shape, average density, and average signal, to account for the stochastic nature of ionization (most track segments have less than mean ionization). The first iteration of the High Atomic Z Astrophysical Radiation Dynamics (HAZARD) simulator generates Galactic cosmic rays via Monte Carlo and propagates those particles through the simulated volume of the detector, including stochastic generation of secondary electrons. The parameters governing charge generation are fit to X-ray source and cosmic ray muon events observed at Goddard’s Detector Characterization Laboratory. We plan to make HAZARD an option in the next iteration of the Roman image simulators for users requiring a more “full physics” treatment of cosmic rays, and we will continue to update it based on cosmic rays observed in flight.
Co-Authors: Emily Koivu, Zac Brutko, Brynn Ayers, and Christopher Hirata
From JWST to Roman: Scalable Machine Learning Pipelines for Automated Brown Dwarf Discovery
By: Julia Sakurako Haynes
Type: Talk
Abstract: The Nancy Grace Roman Space Telescope’s upcoming wide-field surveys are projected to uncover large populations of brown dwarfs, requiring automated pipelines to efficiently isolate these faint targets from dense extragalactic backgrounds. Currently, the James Webb Space Telescope (JWST) has dramatically enhanced our ability to observe faint sources in deep-field surveys, revealing a distant and ancient population of brown dwarfs likely originating from the Milky Way's metal-poor thick disk and halo populations. Identifying these rare objects requires analyzing increasingly large spectral datasets with high efficiency and accuracy. To address this, we trained and evaluated several machine learning (ML) classifiers, including k-nearest neighbors, logistic regression, random forests, and support vector machines on JWST/NIRSpec Prism spectroscopic data to isolate brown dwarfs from predominantly extragalactic sources. We explored performance between the different ML algorithms, variations in the balance between brown dwarfs and non-brown dwarfs of the training set, and the use of principal component analysis (PCA) as an initial data filter. The balanced random forest model achieved the best performance, with an accuracy of 98.43%, brown dwarf recall of 100%, and non-brown dwarf recall of 98.42%. When applied to NEXUS survey data, the model identified 95 brown dwarf candidates out of 3697 spectra, recovering all 37 sources previously confirmed by visual inspection. These results demonstrate the potential of machine learning to accelerate brown dwarf discovery in JWST deep-field surveys, providing a scalable, tested framework directly transferable to processing the massive influx of spectroscopic data expected from Roman's Wide Field Instrument.
Co-Authors: Adam Burgasser, Emma Softich, Marylin Loritsch, Evan Pritchard
Establishing a census of star formation across cosmic time with Roman
By: Sahil Hegde
Type: Talk
Abstract: Though the last three decades of observational searches have begun to formalize a picture of galaxies across time, our understanding of the cosmic evolution of star formation remains incomplete. Galaxy surveys—beginning first in the local universe and now well into the first billion years with JWST—have yielded extensive samples, but systematic uncertainties can be large and interpreting these observations theoretically has proven to be challenging. In turn, exhaustive searches for the first stars near and far have been largely fruitless and the theoretical parameter space remains unconstrained. In this talk I will explore the state of our understanding of star formation from the local universe to the first stars with a focus on where and how Roman is especially well-suited to shed light. First, I will introduce the ‘abcd’ model, a theoretical model for star formation in low metallicity environments, which tracks a galaxy’s star formation from its first stars through reionization. However, such predictions crucially rely on a range of assumptions (e.g., with respect to the low metallicity IMF, star formation efficiency, etc.), exclusively driven by the theoretical models. I will discuss how deep time domain surveys with Roman have the potential to tighten these uncertainties and present forecasts for transient rates achievable exclusively with Roman. Moving beyond the era of the first stars, I will also describe a phenomenological approach to revisit the longstanding disagreement between theoretical and observed galaxy scaling relations across time, such as the star forming main sequence (SFMS). While JWST can begin tackling these problems, its small field of view necessarily makes any results susceptible to cosmic variance, even in the widest surveys. I will conclude by discussing the improved constraints on the SFMS and stellar mass function enabled by Roman, connecting our observations of galaxy populations both near and far.
Co-Authors: Steve Furlanetto
A Foundation Model for Joint Rubin–Roman Survey Processing
By: Shoubaneh Hemmati
Type: Talk
Abstract: Joint Rubin–Roman science will require more than catalog-level cross-matching: detection, astrometric registration, PSF modeling, and photometry must be solved coherently across instruments with different resolutions, bandpasses, PSFs, and noise properties. Classical survey pipelines treat these steps as chained parametric stages, allowing model-dependent residuals from one stage to propagate into the next. We present JAISP, the Joint AI Survey Processing framework, a multi-instrument foundation model that learns a shared pixel-level representation from survey imaging and reuses it across downstream processing tasks. Because each task head reads from the same learned representation rather than from the output of an upstream task, JAISP reduces the sequential propagation of residuals that can arise in chained pipelines. JAISP is built around a self-supervised masked-band autoencoder trained to reconstruct held-out bands from the remaining observations, encouraging the encoder to learn cross-instrument spatial correspondence and chromatic morphology without object-level labels. The trained encoder is then frozen, and lightweight task-specific heads are trained on top of its features for detection, astrometric alignment, PSF estimation, and photometry. Ahead of Roman science operations, we demonstrate the framework on Rubin–Euclid imaging in the ECDFS overlap, a ground–space joint processing problem with the same essential structure as Rubin–Roman, using Euclid VIS/NISP mosaics together with Rubin optical data to address resolution mismatch, PSF heterogeneity, band-dependent morphology, and cross-survey calibration. In our current astrometric alignment test, the learned latent-position head produces a post-correction median cross-instrument residual of approximately 10 mas over nearly 700,000 measurements, demonstrating that the shared representation carries precision-relevant cross-instrument spatial information. The same architecture applies directly to Rubin–Roman once Roman Wide Field Instrument imaging becomes available, providing a unified pipeline for the joint survey era.
Co-Authors: George Helou, Yu Heng Lin, Andreas Faisst, David Shupe, Rachel Akeson, Thomas Greene
Roman color-spectral type relations for ultracool dwarfs
By: Easton Honaker
Type: Poster
Abstract: The Nancy Grace Roman Space Telescope, planned to launch in Fall 2026 and begin survey operations in early 2027, is anticipated to discover thousands of ultracool dwarfs with its exceptional field of view and infrared photometric depth. Compared to other stellar and extragalactic objects, ultracool dwarfs are relatively rare and require specific selection methods to be confidently identified. Color-color cuts are one of the most effective of such selection techniques and applying color-spectral type relations afterwards is a critical step in confirming candidates as ultracool dwarfs. Using both color-color selection and color-spectral type relations will be advantageous to separate the extragalactic and ultracool dwarf “little red dots” that are prevalent in recent deep extragalactic fields. We present color-spectral type relations for field, young, and low metallicity/subdwarf ultracool dwarfs based on recent discoveries from the Backyard Worlds Collaboration and using the Simple Archive. The polynomial fits to the color-spectral type relations will allow for the characterization of ultracool dwarf candidates beyond color-color selection and enable calculations such as photometric distances. With the expected volume of Roman discoveries, tools like color-spectral type relations will help increase candidate evaluation efficiency.
Co-Authors: Easton Honaker (1); Austin Rothermich (2,3,4) ; Jackie Faherty (2); J. Davy Kirkpatrick (5); Marc Kuchner (6); Aaron Meisner (7); Adam Schneider (8); John Gizis (1); Kelle Cruz (2,3); the Backyard Worlds Collaboration (1) University of Delaware; (2) American Museum of Natural History; (3) CUNY Hunter College; (4) CUNY The Graduate Center; (5) Caltech/IPAC; (6) NASA Goddard Flight Center; (7) NSF NOIRLab; (8) USNO, Flagstaff Station
Roman and Some Big Unsolved Problems in Galaxy Formation: a Simulation Perspective
By: Philip F Hopkins
Type: Talk
Abstract: I’ll review some of the dominant uncertainties in predictions and understanding of galaxy formation and evolution, from the point of view of state-of-the-art cosmological simulations. I’ll touch on 5 major areas of uncertainty that can have qualitative, order-of-magnitude effects on galaxy properties, where observations have shown tantalizing hints but our understanding (both theoretical and observational) remains incomplete in critical ways: the role of cosmic rays (both within galaxies and in the CGM/IGM, and their effects on cosmological probes at Mpc scales); “bursty” star formation (shaping the rapid growth of high-z galaxies but also strongly influencing feedback, enrichment patterns, and black hole growth); the origin or ’seeds’ and rapid early growth of supermassive black holes (where they come from and how/where/when/why they grow rapidly at early times); the physical channels of AGN ‘feedback’ (what physics actually couple AGN feedback to their surrounding galaxies, and how can we test it), and unknown ‘dark sector’ physics (what “small-scale problems” of CDM are still problems, and what classes of models produce interesting and testable differences)
Surveying stellar flare rates and properties in a new wavelength regime with GBTDS
By: Ward Howard
Type: Talk
Abstract: Only several dozen stellar flare light curves have been recorded from a handful of stars at infrared wavelengths, preventing any unified understanding of flare mechanisms probed by this wavelength regime or comparison to better-studied regimes. The lack of observations similarly limits our understanding of exoplanet radiation environments and contamination of exoplanet atmospheric characterization. The stochastic occurrence, short timescales, and low infrared fluxes of flares make Roman’s GBTDS the only current or planned facility capable of obtaining robust statistics on infrared flare rates and properties as functions of stellar mass, age, and rotation. We demonstrate the feasibility of a deep learning-assisted flare survey across all 25M sources of F146<27 in the first three high cadence seasons of the GBTDS by presenting survey yields and tests of automated detection. We find the GBTDS will detect 398,000 flares, including 64,000 from sources with detectable stellar rotation periods to infer approximate gyrochronological ages. We show Roman will enable us to (1) determine whether infrared flare rates follow a single or broken-power law distribution, (2) identify differences in the physical processes underlying flare emission at infrared versus optical wavelengths, and (3) compare flare rates and morphologies at 99.7% confidence across a 2D grid of stellar masses and ages from 0.1-1.6 solar masses and 10-10,000 Myr to assess the impacts of spin down. Yields are estimated from the stellar mass, age, and F146W magnitudes of all targets in the GBTDS footprint by interpolating flare rates and detection limits on a (0.08--1.6Msol, 5--9500 Myr) stellar mass-age grid derived from TESS flare rate measurements and JWST light curves. We perform flare injection and recovery tests on simulated light curves using a convolutional real-bogus classifier, measuring true-positive, false-negative, and false-positive rates of 87%, 13% and 1.1% per star, respectively.
Co-Authors: Hank Corbett, UNC-Chapel Hill Adina Feinstein, Michigan State University Zachary Claytor, STScI Guadalupe Tovar Mendoza, Johns Hopkins University Meredith MacGregor, Johns Hopkins University
Cosmological Forecasts for Kinematic Lensing and Multiprobe Strategies in the HLWAS
By: Yu-Hsiu Huang
Type: Poster
Abstract: With the launch of the Nancy Grace Roman Space Telescope approaching, quantifying its cosmological constraining power and optimizing the survey strategy are increasingly time-critical. The medium tier of the High Latitude Wide Area Survey (HLWAS) will deliver overlapping photometric and spectroscopic data across 2400 square degrees. Beyond the standard 3x2pt analysis, this dataset uniquely enables kinematic lensing (KL), a weak lensing (WL) technique that exploits galaxy kinematics to suppress shape noise, the dominant statistical limitation of traditional WL. In this talk, we present cosmological constraints derived from simulated likelihood analyses to evaluate the scientific return of the HLWAS. We benchmark KL against traditional WL, quantify the impact of transitioning from earlier survey definitions to the newly proposed designs, and introduce a KL-based multiprobe analysis that jointly models KL and galaxy clustering. Our results show that this KL multiprobe approach yields the tightest overall constraints on the dark energy equation of state (w0, wa), outperforming any individual probe considered. These findings provide concrete, actionable input to finalize Roman's HLWAS survey configuration and fully realize its dark energy science potential.
Co-Authors: Jiachuan Xu (Northeastern) Elisabeth Krause (U Arizona) Tim Eifler (U Arizona) Vivian Miranda (Stony Brook) Eric Huff (JPL) Pranjal RS (JPL)
Kinematic Weak Lensing with the Roman Space Telescope
By: Eric Huff
Type: Talk
Abstract: Weak gravitational lensing is one of the best probes we have of the dark Universe. Lensing measurements allow us to directly measure the growth of structure over cosmic time, and will be the basis of some of the strongest constraints on the nature of cosmic acceleration coming from Roma's wide-field imaging survey programs.
In this talk, I’ll outline a new way to measure weak lensing using galaxy kinematics to infer the unlensed shapes of galaxies. This lets us control for almost all of the noise in weak lensing measurements, while largely eliminating the currently dominant sources of systematic error. I’ll describe the method, talk about the recent first detection, and outline the current program for carrying out kinematic lensing measurements with the Roman telescope’s spectroscopic survey. Even with the small fraction of Roman galaxies for which kinematic measurements will be possible, this method offers a factor of 3 gain in dark energy constraints relative to the primary Roman lensing surveys.
Co-Authors: Tim Eifler, Elisabeth Krause, Pranjal R.S., Jiachuan Xu, Yu-Hsiu Huang, Eiichi Egami, Spencer Everett
A saturated photometry approach for asteroseismology in the Galactic Bulge Time Domain Survey
By: Lauryn Michelle James
Type: Poster
Abstract: The Nancy Grace Roman Space Telescope Wide Field Instrument (WFI) will be used to conduct the Galactic Bulge Time Domain Survey (GBTDS), observing millions of stars with a cadence that is suitable for asteroseismology of red giants. An important challenge will be to recover WFI F146 photometry in the saturated regime, particularly for asteroseismology of giants but also for the core survey microlensing science goals — both of which require time-resolved relative photometry. We characterize here the conditions under which the combined effects of saturation, crowding, and persistence exceed tolerable thresholds for time domain photometry, with an eye toward asteroseismology of giants in the bulge. The operating principle of the proposed saturated photometry approach is to extract relative flux from unsaturated wings of the PSF in L1 resultant-level data using halophot – an empirical PSF photometry method from the literature. To test this approach, we describe a simulation pipeline based on romanisim and romancal. A GBTDS observation strategy is adopted, injecting sources drawn from the Galaxia stellar population synthesis tool. The resulting light curves enable saturated photometry for red giant asteroseismology of the bulge in the limit of ideal crowding corrections. We also explore an outlier rejection technique to reduce flux contamination from crowding in the presence of the combined effects of a nominal dithering pattern and persistence. We discuss ongoing improvements to the resulting light curve simulation tool and saturated photometry pipeline, including crowding corrections using PSF photometry.
Co-Authors: The Roman Asteroseismology Wide Field Science Team
Enabling Time-Domain Science for Roman with RAPID
By: Jacob Jencson
Type: Talk
Abstract: The Nancy Grace Roman Space Telescope is set to transform our view of the dynamic infrared sky. RAPID (Roman Alerts Promptly from Image Differencing) is a Project Infrastructure Team designed to enable a wide suite of transient and variable astrophysics, identified as a high priority in the Astro2020 Decadal Survey. Roman's wide field, high spatial resolution, and near-infrared sensitivity will open new discovery space in numerous areas, including dust-enshrouded transients, kilonovae, high-redshift supernovae, and eruptive and pulsating variable stars. Based at Caltech and drawing on experience with the Zwicky Transient Facility, the RAPID team will provide four core services to the Roman community: rapid, low-latency image differencing of every new Roman image against a reference; a prompt public alert stream of all Roman transient and variable candidates; source-matched light curve files for every identified candidate; and an on-demand forced-photometry service for the photometric history at any observed sky location. In this talk, I will highlight science cases enabled by RAPID and provide updates on our pipeline developments as Roman's launch approaches.
Co-Authors: The RAPID Team
The Emerging Black Hole Mass Function in the High-Redshift Universe
By: Junehyoung Jeon
Type: Talk
Abstract: An abundant population of supermassive black holes (SMBHs) has been observed to be already in place during the first few hundred million years of cosmic history. Most of them appear overmassive relative to the stellar mass in their host systems, challenging models of early black hole seeding and growth. Multiple pathways exist to explain their formation, including heavy seeds formed from direct collapse/supermassive stars or sustained super-Eddington accretion onto light stellar remnant seeds. We use the semi-analytical code A-SLOTH to predict the emerging SMBH mass function under physically motivated models for both light and heavy seed formation, to be compared with upcoming ultra-deep JWST and wide-field Roman surveys. We find that both pathways can reproduce observations at z ∼ 5 − 6, but have distinct features at higher redshifts. Specifically, Roman observations of SMBH populations at z~10 have the potential to constrain the fraction of efficiently accreting (super-Eddington) SMBHs through blank field surveys. Such observations will provide key insights to understand the process of SMBH formation and evolution during the emergence of the first galaxies. We emphasize the great promise of possible SMBH detections at z ≳ 10 with future Roman observations to break the degeneracy between light- and heavy-seed models.
Co-Authors: Boyuan Liu, Anthony J. Taylor, Vasily Kokorev, John Chisholm, Dale D. Kocevski, Steven L. Finkelstein, Volker Bromm
Effective EoS in Interacting Dark Sector Models: Constraints from Roman Supernovae
By: Hussain Ahmed Khan
Type: Poster
Abstract: The cosmological constant, and the assumption of effectively pressureless matter in the ΛCDM
framework, while arguably observationally successful, obscures the rich physics that may lie hidden
in the dark sector. One of the ways extensions of this framework can be made is if we allow
interactions of matter–dark energy. In our work we consider a class of Matter-Dark Energy
interacting models. We will be focusing on the interactions between matter and dark energy
specifically, in which an exchange of energy-momentum between these two sectors gives rise to a
modification of the standard matter evolution, which can be interpreted as an effective equation-ofstate like behavior, without introducing a fundamental non-zero pressure for matter. We will examine
two interacting scenarios: Q = 3Hξ (ρde + ρm) and Q = 3Hξρde, for various values of the ξ parameter.
We use these modified models to generate simulated Type Ia supernovae datasets using SNANA,
combining low redshift samples from DEBASS and high redshift projections from the Nancy Grace
Roman Telescope. In order to determine whether these interactions may be absorbed by the standard
parameterization of wCDM and w0waCDM, we fit the simulated data sets to these models.
We explore how these interaction-induced modifications manifest in the expansion history and the
resulting supernova observables, and assess their distinguishability within standard cosmological
parameterizations.
Co-Authors: B. M. Rose
A Roman WFI Grism Census of Ionizing Efficiency and Bursty Star Formation Across Cosmic Noon
By: Keunho Kim
Type: Talk
Abstract: The sources powering cosmic reionization remain one of the central open questions in galaxy evolution, with major uncertainty surrounding which galaxy populations dominate the ionizing photon budget and how their star formation histories regulate ionizing photon escape and production. I will present new constraints on the ionizing photon production efficiency (ξion) for low-mass galaxies down to (log(M* / M⊙) ≃ 7.5) at (0.5 < z < 2.3), utilizing deep JWST/NIRISS slitless spectroscopy and strong-lensing magnification of lensed galaxies. Based on a sample of 148 galaxies with robust detections in both H-alpha and the UV continuum (SNR > 3 for each), we find a gradual increase in the median (ξion) with increasing redshift across this range, confirming previous results and extending them down to (z=0.5). We also find a significant correlation between (ξion) and UV continuum slope, with bluer galaxies exhibiting higher (ξion), likely reflecting the presence of young, massive stars formed during recent bursty star formation episodes. Moreover, our high-redshift (1.5 < z < 2.3) low-mass galaxies show an (≈ 0.2) dex increase in median (ξion) compared to more massive systems (log(M* / M⊙) > 9), suggesting that low-mass galaxies are more efficient ionizing photon producers than their massive counterparts.
In addition, I will discuss how the Roman WFI grism will transform these studies from small lensed samples into a statistical demographic census of ionizing photon production and bursty star formation across cosmic noon (0.5 < z < 2). Roman’s combination of wide-area near-infrared slitless spectroscopy and sensitivity, together with future wide-field UV imaging surveys such as UVEX, will enable joint measurements of H-alpha and UV continuum emission for unprecedented galaxy samples spanning a broad range of environments and stellar masses.
Co-Authors: Alavi, Anahita ; Watson, Peter; Colbert, James ; Mehta, Vihang ; Morishita, Takahiro ; Siana, Brian ; Sattari, Zahra ; Teplitz, Harry ; Wang, Xin
Little Red Dots and the Rise of Obscured AGN in the Early Universe
By: Dale Kocevski
Type: Talk
Abstract: One of the more surprising results from JWST has been the discovery of a large population of red, broad-line AGN known as “little red dots” (LRDs), which are ubiquitous in the early Universe (4
Towards a Unified Machine Learning Model for Time-Domain Alert Classification Across Roman Surveys
By: Aayush Kuloor
Type: Poster
Abstract: Maximizing the scientific returns from the Nancy Grace Roman Space Telescope’s time-domain surveys, including the High Latitude Time Domain Survey (HLTDS), the Galactic Bulge Time Domain Survey (GBTDS), and the Galactic Plane Survey (GPS), requires a rapid, reliable alert classification pipeline. However, with an estimated 12.5 billion alerts anticipated over the five-year mission, human verification is rendered impossible, making fully automated pipelines a necessity. A major challenge in developing these pipelines is the extreme variations in stellar density, exposure time, depth, and observing cadence across Roman’s planned surveys. Given the sheer number of distinct machine learning models required to cover every combination of surveys and filters, developing individual classifiers is highly impractical. Therefore, a robust, generalized solution is essential.
This work outlines a unified machine learning model designed to generalize across the diverse environments encountered in Roman time-domain surveys. By synthesizing simulated survey data spanning from extragalactic HLTDS-like observations to dense GBTDS fields, we have developed a unified base model built upon a transformer architecture. Crucially, we utilize attention mechanisms and continuous time embeddings, allowing the model to learn generalized temporal dynamics independent of the heterogeneous sampling cadences of specific surveys.
Building upon a successfully developed engine designed to filter instrumental artifacts (Real/Bogus), we have adapted elements of this architecture to directly categorize genuine astrophysical sources (Real/Real). This enables the rapid classification of distinct phenomena, such as astrophysical transients and microlensing events. Developed as part of the Roman Alerts Promptly from Image Differencing (RAPID) pipeline, these classifiers can be directly incorporated into alert brokers, preparing the Roman alert ecosystem to begin making robust discoveries upon the start of operations later this year.
Co-Authors: Ashish Mahabal, Jacob Jencson, and the RAPID Team
Peering into the streams of external galaxies using Roman in the context of the near-far technique
By: Aritra Kundu
Type: Talk
Abstract: The ``near-far'' approach to studying reionization leverages the star formation histories of nearby small galaxies in the Milky Way (MW) or Local Group (LG), derived from resolved photometry, to infer the low-mass/faint end of the stellar mass functions (SMFs) or the ultraviolet luminosity functions (UVLFs) of high-redshift galaxies beyond JWST detection limits. The Roman Space Telescope will provide us with the resolved-star observations necessary to perform this reconstruction in more distant external galaxies for the first time, even for galaxies that are now tidal streams. Previous theoretical works used to validate this technique considered only intact low-mass galaxies in the MW and LG, neglecting disrupted galaxies such as stellar streams and considering only individual systems. I show, using the FIRE-2 simulations, that disrupted galaxies contribute up to ~50% of the total stellar mass budget of the proto-MW/LG enivironment at z = 6 - 9. Neglecting them thus underestimates the contribution of relatively brighter galaxies to the reionization-era UV luminosity density by a significant and mass-dependent factor. Including these disrupted galaxies improves the normalization of the recovered SMFs/UVLFs by factors of ~2 to 3 and reduces the halo-to-halo variation in the slope by ~20-40%. Using synthetic Roman observations of our simulated galaxies, I will quantify how well we can recover the faint-end slope by using only substructures detectable by Roman in external host galaxies at different distances, thus extending the volume of space accessible to the near-far technique by orders of magnitude.
Co-Authors: Robyn Sanderson, Adam Lidz
Synergizing SPHEREx and Roman for Galactic Plane Background and ISM Studies
By: WANGGI LIM
Type: Talk
Abstract: While the Roman Space Telescope’s Galactic Plane Survey aims to map billions of stars, the complex diffuse Galactic background poses a major challenge to photometric precision and detection depth in crowded fields. We present a synergistic framework that adapts an existing Galactic diffuse background model built upon SPHEREx’s all sky spectral imagery from 0.75 to 5 microns across 102 channels to upcoming Roman WFI observations from 1 to 2.3 microns. By synthesizing highly accurate 2D background templates tailored to Roman’s dedicated bandpasses, our model drastically improves point source extraction and completeness limits in dense stellar regions. Furthermore, analyzing the residuals between our diffuse model and Roman’s ultra high spatial resolution of 0.11 arcseconds per pixel allows us to isolate and map previously unresolved sub parsec scale dust filaments and interstellar medium microstructures. This methodological bridge between SPHEREx’s macroscopic spectral fidelity and Roman’s microscopic spatial resolution demonstrates how cross mission joint analysis can maximize the scientific returns of next generation infrared cosmic surveys.
Co-Authors: Jae Hwan Kang (Caltech) Keunho Kim (Caltech / IPAC)
Stellar-mass black holes with the Galactic Bulge Time Domain Survey
By: Casey Lam
Type: Talk
Abstract: Although there are expected to be 100 million stellar-mass black holes in our Galaxy, to date only a few dozen have been confirmed. Furthermore, this sample is highly biased: while >99% of Galactic black holes are expected to be isolated or in detached non-interacting binaries, the majority of confirmed detections are found in highly interacting close X-ray binaries. This unrepresentative sample prevents us from understanding the true population properties of black holes, such as their mass distribution, binary fraction, and natal kick velocities. The Roman Galactic Bulge Time Domain Survey will solve this problem by allowing us to build a representative sample of isolated and detached binary black holes. Thanks to the GBTDS's precise astrometry, Roman will characterize ~100 such black holes via astrometric microlensing and Keplerian reflex motion. This sample of black holes will provide the observational constraints needed to understand the evolution and death of massive stars, constrain the physics of binary interactions, mergers, and disruptions, and contextualize the population of highly interacting black holes found in X-ray binaries and gravitational wave mergers.
Inferring Planet Occurrence Rates in Different Galactic Populations: Connecting K2 and Roman
By: Matthew Lastovka
Type: Poster
Abstract: The Galaxy is composed of several distinct stellar populations, including the thin disk, thick disk, bulge, and halo, each characterized by different distributions of ages, chemical compositions, and dynamical histories. Therefore, these galactic substructures provide a powerful laboratory for studying how stellar properties influence planet formation and evolution. The best current exoplanet survey capable of performing these calculations is the K2 survey. I combine a K2 planet catalog and stellar chemical abundances from the APOGEE survey to compare exoplanet occurrence rates in the thin and thick disks. I present preliminary results that suggest the thick disk is slightly depleted in planets relative to the thick disk. The difference persists even after accounting for the change in planet occurrence resulting from the metallicity differences between the disks. However, these results are limited by K2’s relatively small sample size in the thick disk. While previous occurrence rate calculations have suffered due to small sample size beyond the thin disk, the Roman Space Telescope’s Galactic Bulge Time Domain Survey (GBTDS) will provide an unprecedented opportunity to study planet occurrence across galactic populations. Predictions estimate that we will detect ~100,000 transiting exoplanets, including thousands in the thick disk and bulge, allowing us to probe demographics across populations in unprecedented detail. I present a framework that utilizes galactic population synthesis models to measure occurrence rates across different galactic populations using Roman data.
Co-Authors: Thomas Barclay (NASA Goddard Space Flight Center), Robby Wilson (University of Maryland; NASA Goddard Space Flight Center), Emily Gilbert (Caltech/IPAC-NASA Exoplanet Science Institute)
SN2025wny: A Hydrogen-poor Superluminous Supernova at z ≈ 2
By: Maggie Li
Type: Poster
Abstract: The properties of stars in the early universe remain poorly understood. At high redshift, lower metallicities are expected to significantly affect stellar evolution and end-of-life outcomes. Hydrogen-poor superluminous supernovae (SLSNe-I), which preferentially occur in low-metallicity high-redshift galaxies and are 10-100 times brighter than classical core-collapse supernovae (CCSNe), provide a powerful probe of these early stellar populations and their explosion physics. However, SLSNe comprise less than 1% of all SN discoveries and current studies are strongly biased toward local (z < 1) events, leaving their physics and possible cosmic evolution poorly constrained. Here, we present SN2025wny, a gravitationally lensed SLSN-I at z = 2.011 and the most distant spectroscopically confirmed SLSN to date. We present extensive spectroscopic and photometric observations spanning the rest-frame UV to NIR throughout the photospheric phase, alongside an analysis of its spectral evolution, light-curve properties, and ejecta composition. Comparisons to well-studied local SLSNe-I show that SN2025wny differs significantly from canonical members of its class, exhibiting an extreme intrinsic luminosity, unusual spectral evolution, and rest-frame UV and optical features not commonly observed in nearby events. These observations provide initial constraints on the physics of high-z SLSNe-I, and together with continued monitoring into the nebular phase, establish the first robust spectral templates of a high-z SLSN-I, providing an essential reference for identifying and analyzing high-z SLSNe in the upcoming Rubin and Roman era.
Co-Authors: Lin Yan, Joel Johansson, Ariel Goobar, Yujing Qin, Suhail Dhawan, Alice Townsend, Mansi Kasliwal, Steve Schulze, Daniel Perley, Ragnhild Lunnan, Jesper Sollerman
From Cold Filaments to Hot Nodes: Protocluster Assembly Pathways at Cosmic Noon with Roman
By: Yongming Liang
Type: Poster
Abstract: Massive protoclusters at z~2 are the progenitors of today's richest galaxy clusters, yet systems with comparable descendant halo masses are observed in very different baryonic states: some embedded in cold IGM filaments, some hosting giant Lyα nebulae, and others already showing starburst-heated CGM or nascent thermal ICM. Does this diversity trace a single evolutionary sequence sampled at different stages, or genuinely distinct assembly pathways? I will discuss a Roman-era strategy for addressing this question by jointly mapping galaxy populations, halo assembly, cosmic-web geometry, and multiphase gas across the full extent of extreme overdensities. Roman/WFI imaging and G150 slitless spectroscopy enable homogeneous measurements of metallicity and ionization, quiescent populations, AGN incidence, and rest-UV/optical morphology. Density-field reconstruction, supported by lensing constraints where feasible, can then connect these observables to halo mass and cosmic-web structure. Archival multiphase gas tracers can be registered onto this galaxy-defined structural frame to test whether enrichment, quenching, and feedback signatures vary coherently along a cold-to-hot gas-phase sequence, or instead reveal multiple tracks coexisting at fixed descendant mass. Well-studied extreme overdensities with rich archival gas diagnostics provide high-contrast anchors at the rarest density peaks, while public Roman CCS/HLWAS Deep fields extend the same diagnostics over wide blank-field areas at matched depth, enabling blind protocluster discovery and statistical control. Together, these data support a population-level test of how the most massive halos assemble at cosmic noon.
Co-Authors: Yunjing Wu, Masami Ouchi, Tadayuki Kodama, Zheng Cai, Rhythm Shimakawa, Sebastiano Cantalupo, Tao Wang, Nobunari Kashikawa, Xiaohui Fan, Yongda Zhu, Kentaro Nagamine, Hidenobu Yajima, Ken Osato, Guochao Sun, Mingyu Li, Fengwu Sun, Jiachuan Xu et al.
The impact of Roman on accurate X-ray source classification
By: Yichao Lin
Type: Poster
Abstract: The Nancy Grace Roman Space Telescope will have a significant impact on astrophysics outside of the optical-NIR domain. Currently, there are hundreds of thousands of catalogued X-ray sources with accurate (~<1'') positions. These are dominated by faint objects whose classifications often require considering associations (or establishing the lack thereof) with optical-IR counterparts down to very deep detection limits. Among these low-luminosity X-ray sources are quiescent black hole low mass X-ray binaries (such as those recently associated with PeV LHAASO sources), millisecond pulsars in binaries (which may be responsible for e+/e- annihilation line emission near the galactic center), and ultra-compact X-ray binaries (which are expected to be the dominant galactic source of Gravitational Waves for LISA). The upcoming Roman Galactic Plane Survey will provide an order-of-magnitude superior angular resolution and improved survey depth compared to existing wide-field surveys, which is critical for the detection of faint multiwavelength counterparts, particularly in the complex and often highly absorbed Galactic plane environment. Here, we discuss and demonstrate with examples the potential impact of Roman on the classification of unknown X-ray sources.
Co-Authors: Oleg Kargaltsev, Jeremy Hare
Synergizing Roman and Line Intensity Mapping: Forecasting Effects of Bursty Star Formation in The Early Universe
By: Lun-Jun Liu
Type: Talk
Abstract: Recent JWST observations have revealed a surprising overabundance of UV-luminous galaxies during the Epoch of Reionization (EoR) and Cosmic Dawn. A critical open question is whether this excess is driven by realistic astrophysical mechanisms, such as highly stochastic star formation histories and feedback-free enhanced star formation efficiencies, or simply cosmic variance within narrow JWST fields. The Nancy Grace Roman Space Telescope’s unprecedented wide-field capabilities are uniquely positioned to break this degeneracy. In this talk, we present a novel simulation framework designed to investigate the effects of early bursty star formation on galaxy statistics measured by Roman’s High-Latitude Survey. By painting parameterized bursty star formation effects onto dark matter halo catalogs within simulated cosmic volumes, we model the spatial distribution of bright early galaxies while effectively capturing the underlying statistics of the populations. Through this framework, we provide forecasts for Roman, predicting the impact of burstiness on the number density and UV luminosity function of Lyman-break galaxies up to z ~ 15. We will utilize this framework to forward-model the clustering of these early galaxies and their luminosity-weighted bias. This approach will further enable multi-line luminosity function constraints and galaxy x Line Intensity Mapping (LIM) cross-correlations, exploring powerful synergies between Roman and recent LIM experiments. Anchoring aggregate LIM signals to Roman’s wide-field galaxy distributions will also provide critical insights into early bursty star formation, the EoR timeline, and the topological growth of ionized bubbles.
Co-Authors: Adam Lidz, Guochao (Jason) Sun, Aritra Kundu, Jordan Mirocha, Tzu-Ching Chang
Little Red Dots in the Roman Grism Era: Census and Black Hole Mass Constraints from Spectroscopic Variability
By: Zhaoran Liu
Type: Talk
Abstract: Little Red Dots are among the most puzzling sources in the early universe, possibly hosting the long-sought seed black holes.
Growing evidence suggests their central engines may be buried in gas envelopes, but their physical nature remains debated, and resolving it requires both large samples and variability constraints.
In this talk, I will present the JWST TWINKLE program, the first systematic LRD variability study using JWST/NIRCam grism spectroscopy. Across 27 broad-line emitters at z = 3.9-6.8 monitored over ~200 days in the rest frame, TWINKLE finds no variability, inconsistent with traditional AGN expectations and may point toward low-mass black holes accreting in the super-Eddington regime.
Roman's grism opens a complementary window at z ~ 0.5-2, bridging the redshift gap between JWST's high-z LRDs and the local population. Wide-area Roman grism surveys, such as HLTDS, combined with the longer time baselines naturally enabled by Roman's survey cadence, will provide both larger samples and tighter variability-based constraints on LRD central engine physics across cosmic time.
Co-Authors: Naidu, Rohan P., Simcoe, Robert A.
Finding Free-Floating Black Holes in Abundance with Roman
By: Jessica Lu
Type: Talk
Abstract: The landscape for studies of stellar-mass black hole origins, evolution, and demographics has expanded dramatically not only with the detection of gravitational waves; but also the explosion of EM photometric and astrometric time domain surveys. Gravitational microlensing is particularly powerful as it is the only means of finding and weighing isolated black holes across the mass spectrum. Time-domain surveys from Roman, particularly the Roman Galactic Bulge Time Domain Survey, are particularly valuable as it provides simultaneous photometry and astrometry of crowded stellar fields where microlensing events are more common. We will present population simulations, mock datasets, and model fits that show how sensitive the Roman GBTDS microlens sample is to the black hole mass function, binary fraction, and velocity distribution. Roman will likely expand the sample of known Milky Way black holes by 100x in the coming decade. This sample of Milky Way black holes, both isolated and in binaries, will be invaluable when combined with the black hole sample from gravitational wave experiments and astrometric wobbling black hole + star binaries detected with Gaia and other astrometric experiments.
Co-Authors: Howard Isaacson, Macy Huston, Natasha Abrams, Casey Lam, Sean Terry, Dex Bhadra, Alina Hussain, Madelyn Rahimi, Sage Remulla, Caitlin Begbie, Tianjin Lao, Shep Brooke, Riley Patlak, Matt Hosek, Hannah Gulick
Arandu: A Specialized Alert Broker for Roman Time-Domain Science
By: Ashish Mahabal
Type: Talk
Abstract: The Nancy Grace Roman Space Telescope will open a qualitatively new regime for time-domain astronomy: wide-field infrared imaging from space, high source densities, heterogeneous cadences, and a combination of Galactic and extragalactic science drivers. Existing alert-brokers developed for ZTF/LSST, including Fink, ALeRCE, Babamul, have demonstrated the power of real-time filtering, and classification. Roman, however, will require a broker architecture tuned to its own survey modes, constraints, and science priorities.
We present the concept for Arandu, an alert broker for Roman that will be run from the Brazilian Center for Research in Physics (CBPF), located in Rio de Janeiro, Brazil (PI: Clécio Bom). Drawing on lessons learned from current brokers, Arandu will address Roman-specific challenges, including the dense Galactic Bulge Time-Domain Survey, the heterogeneous High Latitude Time Domain Survey, and other programs that will generate variable and transient sources. The broker will ingest Roman alerts and associated metadata, support science-driven filtering and ranking, and enrich alerts with contextual information from surveys such as Rubin, ZTF, Euclid, Gaia etc.
A central goal of Arandu is to make Roman time-domain discovery more actionable. This includes support for survey-aware filtering, prioritization of rare or rapidly evolving sources, and pathways for later classification, anomaly detection, and human-in-the-loop feedback. In the early Roman era of sparse labels and imperfect simulations, Arandu can also provide a practical framework for active learning, validation, and community-driven refinement of broker products. We describe the initial design goals, anticipated data flows, and opportunities for collaboration in building a specialized Roman broker while interoperating with the broader time-domain ecosystem. In addition, we present the computational infrastructure in CBPF dedicated to the operation of Arandu.
Arandu will serve as a bridge between Roman’s unprecedented time-domain data stream and the scientific community, helping transform alerts into prioritized, contextualized, and interpretable discovery pathways.
Co-Authors: Clécio Bom, Gabriel Teixeira, Karan Gandhi, Jacob Jencson, Mansi Kasliwal, Aayush Kuloor, Ryan Lau
Roman and the Future of Globular Cluster Asteroseismology
By: Madeleine Louise McKenzie
Type: Talk
Abstract: Roman will provide an unmatched combination of wide-field coverage, high spatial resolution, sensitivity, and time-domain capability, bringing dense stellar systems into the era of population-scale time-domain astrophysics. In globular clusters, Roman will detect oscillating red giants in both crowded cores and extended cluster outskirts, transforming cluster asteroseismology from small-sample studies into a population-level probe of stellar evolution. When combined with multi-band photometry, large spectroscopic surveys, and targeted high-precision spectroscopic follow-up, these data will link stellar masses, chemical abundances, and cluster structure to address open questions in red giant mass loss, globular cluster formation, and stellar evolution in old, metal-poor environments. The same observations would also enable searches for transiting planets at low metallicity and identify pulsators, binaries, and compact-source counterparts. In this talk, I will highlight the emerging synergy between cluster asteroseismology and stellar spectroscopy as a foundation for future Roman campaigns, and present 47 Tuc as an ideal pilot field: it is nearby, extensively characterized, well separated from the Galactic plane, and projected against the Small Magellanic Cloud, making it a high-impact benchmark for Roman studies of stellar populations.
Co-Authors: Dr. Madeline Howell (OSU), Dr. Alison Crisp (OSU), Dr. Geoffrey Mo (Caltech/Carnegie)
Studying galaxy evolution with JWST slitless spectroscopy: forecasting expectations for Roman's HLWAS
By: Vihang Mehta
Type: Talk
Abstract: Over the past decade, near-IR slitless spectroscopy has been pivotal in advancing our understanding of galaxy evolution across cosmic time, providing crucial spectroscopic coverage for statistically significant and unbiased galaxy samples over wide redshift ranges, particularly in the low-mass regime. In this talk, I will summarize results from PASSAGE, a large JWST/NIRISS pure-parallel grism survey. PASSAGE grants access to the key strong emission lines in galaxies at the cosmic noon (z~1-3), which allows us to study their star-formation, gas-phase metallicity, nebular dust properties, and ionization states. Additionally, it also opens a window to the Epoch of Reionization enabling us to understand the properties of the galaxies that reionized the universe. Lastly, I will also describe an upcoming program that uses the newly available JWST/MIRI slitless spectroscopy and how it unlocks an exciting new parameter space for studying PAH emission in galaxies at cosmic noon. The Nancy Grace Roman Space Telescope, launching eminently, is set to revolutionize this field with its High Latitude Wide-Area Survey (HLWAS) and I will discuss how we expect Roman to complement and enhance our current analyses as well as improve our knowledge of the physical processes governing galaxy growth and evolution.
Constraining Type Ia SNe Volumetric Rates and Progenitor Systems with the Roman Space Telescope
By: Cole Meldorf
Type: Talk
Abstract: Despite being one of the most well-studied cosmological probes, a major uncertainty underlies the utility of Type Ia Supernovae (SNe Ia) as a standardizable candle: their progenitor system is poorly understood. In the absence of a direct detection, previous research on the Delay Time Distribution (DTD) of SNe Ia has not been able to statistically conclude if the SNe Ia progenitor is singly-degenerate (SD), doubly-degenerate (DD), or a mix of both. This is largely due to the lack of high redshift data; to measure the DTD requires an accurate measurement of the Volumetric SNe Ia Rate (VSNR), and even combining all available data at z > 1, the statistical uncertainty on the VSNR is still significantly over 10%. This makes discerning between different progenitor models with current data intractable. We use simulated data for the upcoming High-Latitude Time-Domain Survey (HLTDS) to determine what constraints the Roman Space Telescope will place on the high redshift VSNR, and how these constraints translate into restrictions on the possible SNe Ia progenitors. We utilize a method requiring no spectroscopic data to take advantage of the full statistical power offered by the HLTDS dataset. We find that using the HLTDS data will increase precision on the z > 1 VSNR by a factor of 7.7 while having five times finer binning. Assuming a simple model for the DTD, we find that Roman will improve the precision on the fraction of prompt versus delayed SNe Ia by a factor of 7. Depending on the exact fraction measured, these tight constraints could confirm or refute if SNe Ia have a solely SD progenitor, potentially reducing systematics in SNe Ia distance measurements.
Co-Authors: Masao Sako, Richard Kessler
A possible Exozodi survey of HWO targets with the Roman Coronagraph
By: Bertrand Mennesson
Type: Talk
Abstract: Despite its “tech demo” status, the Roman Coronagraph Instrument (RCI) expected performance of at least ~1E-8 instrumental contrast is sufficient to yield exciting exoplanetary system science during its nominal 90 days of observing time.
Of particular relevance to HWO will be RCI’s potential to image terrestrial zone dust (“exozodi”) at visible wavelengths for the first time. If too much exozodiacal dust is present in a system, HWO may be unable to detect an Earth analog. At the moment, the LBTI HOSTS thermal infrared interferometry survey provides the best limits on exozodi around nearby stars. But typical measurements per individual star range at the 100-200 solar zodis level, and an extrapolation from thermal IR to visible is still required. RCI provides an opportunity to empirically verify these predictions, as well as constrain the inclination and PA of the planetary disk.
Furthermore, HWO yield predictions often assume exozodi dust disks are simply a “nuisance” foreground source of photon noise that can be well-subtracted to reveal any exoplanets within. However, zodiacal dust has clumpy resonant structures which can complicate planet detection. Additionally, residual starlight leakage can generate features that can be confused for disk structures. To ground the HWO yield model assumptions, we need to test our ability to both recover the exozodi surface brightness spatial distribution from the RCI observations and also to remove exozodi to recover embedded exoplanets.
Consequently, we have planned exozodi observations of a handful of high priority HWO targets in FY 27, early in the Roman Coronagraph Observation Phase that will 1) provide exozodi limits on potential HWO targets and 2) help us understand the extrapolation from MIR zodi detection to scattered light brightness.
Co-Authors: B. Mennesson (JPL), J. Hom (University of Arizona), J. Llop-Sayson (JPL), V. Bailey (JPL), Schuyler Wolff (University of Arizona) and Karl Stapelfeldt (JPL)
Searching for transiting circumplanetary disks in the Galaxy
By: Tiffany Meshkat
Type: Talk
Abstract: Circumplanetary disks are a fundamental but poorly understood phase of giant planet formation. Only a handful of candidates have been identified so far, all found serendipitously. We present our plans for a systematic search for circumplanetary disk (exoring) transits using the Roman Galactic Bulge Time Domain Survey (GBTDS), which monitors ~60 million stars at ~12- minute cadence over ~70 continuous days per season. We will implement an automated sliding- window detection framework to identify asymmetric transit signatures, with dual-band validation. We will also perform PSF-fitting photometry on a per-target basis to optimize deblending in the crowded bulge fields, supplementing the project-generated difference imaging light curves. Confirmed candidates will be characterized through detailed light curve modeling to constrain disk geometry, gaps, opacity, and substructure, providing direct insight into the physical properties of compact circumplanetary material. Our yield estimate ranges from 60 to 800 new detections depending on disk lifetime assumptions. These detections will advance our understanding of circumplanetary material from isolated curiosities toward a characterizable population, directly informing models of giant planet formation and the origin of ring systems. A non-detection, however unlikely, will place the first meaningful upper limit on exoring occurrence rates around field-age stars.
Co-Authors: Matthew Kenworthy (Leiden University), David Ciardi (Caltech/IPAC), Sammy Mokkenstorm (Leiden University)
Diffmetal: painting self-consistent evolutionary tracks of galaxy metallicity and abundance ratios onto halo merger trees across mass and redshift
By: Kaustav Mitra
Type: Poster
Abstract: We present Diffmetal, an empirical model for the metallicity and abundance ratio evolution of galaxies across cosmic time. Diffmetal separately tracks type-Ia and type-II supernovae enrichment pathways. The redshift evolution is governed by star formation history, supernova feedback and baryon cycle between interstellar medium and circumgalactic medium, and dilution via inflow of pristine gas. It is built on two foundational pillars: a parametric model for halo mass assembly (Diffmah), and an empirical model for halo assembly-informed star formation history (Diffstar). This physics-driven correlated evolution of galaxies and their host halos enable Diffmetal to uniquely construct metallicity evolution tracks from simulated or Monte Carlo-generated halo merger trees across mass and redshift. Diffmetal is a physical model written fully in the AI/ML framework called JAX, which allows us to optimize our physical parameters with the same computationally efficient algorithms that are used to train modern neural networks, without sacrificing traditional methods of uncertainty quantification. We will highlight results from our ongoing effort to use Diffmetal to analyze ongoing and upcoming cosmology surveys such as DESI, Rubin, and Roman to extract galaxy formation physics, specifically about star formation, supernova feedback and the baryon cycle.
Co-Authors: Andrew P. Hearin, Andrew Benson, Andrew Robertson
Multi-survey joint analysis of galaxy formation physics across the cosmic history
By: Kaustav Mitra
Type: Poster
Abstract: We present Diffsky, an empirical modelling of spectral energy distributions (SEDs) of galaxies linked to the co-evolution of their host dark matter halos. The modeling framework populates simulated or Monte Carlo halo merger trees with galaxy SEDs across lightcones spanning any redshift range. I will highlight the ongoing work at Argonne National Lab to accelerate lightcone generation and simultaneous modelling of multi-wavelength, multi-survey datasets. In particular, I will introduce DisCoWebS, Diffsky fit to Cosmos-Web and SDSS, to jointly constrain galaxy formation physics across 12 billion years of cosmic history. These constraints enable the creation of highly realistic OpenUniverse mock catalogs, essential for validation pipelines of current and upcoming cosmological surveys such as DESI, Rubin, Roman, and Euclid. The goal of the presentation is to highlight the power of Diffsky and showcase a glimpse of what can be achieved by applying the same empirical modelling tools to Roman data.
Co-Authors: Andrew P. Hearin, Natália V. N. Rodrigues, Kumail Zaidi, Alex Alarcon, Gillian Beltz-Mohrmann, Georgios Zacharegkas, Matthew R. Becker.
Polar-Structure Galaxies Across Time: A New Frontier for Roman
By: Aleksandr Mosenkov
Type: Poster
Abstract: Galaxies hosting strongly misaligned stellar and gaseous components provide a unique window into external accretion and merger-driven assembly, yet polar-structure galaxies (PSGs) have long been considered rare and largely confined to the local Universe. We present a synthesis of recent results that dramatically expand the known PSG population across cosmic time, from a large, homogeneous low-redshift sample identified in DESI imaging (COUGS–DESI) to intermediate- and high-redshift candidates discovered in JWST CEERS and Euclid Q1 data, where robust polar structures are detected out to z~1. Together, these studies establish PSGs as a measurable population over a wide redshift range and enable initial constraints on the frequency and nature of kinematically decoupled structures. Looking ahead, the Nancy Grace Roman Space Telescope, with its combination of wide-area, high-resolution near-infrared imaging, will be transformative by enabling the discovery of large, statistically significant samples of PSG candidates, opening a new frontier for studying rare merger- and accretion-driven phenomena across cosmic time.
Co-Authors: Seneca Bahr, Jacob Guerrette
Microlensing with Roman to find primordial black holes (PBHs)
By: Jeremy Mould
Type: Talk
Abstract: It’s fifty years since Stephen Hawking laid out the physics of PBHs and fifty years since the cold dark matter paradigm became the standard model for the formation of structure. Theoretically, PBHs range from the Planck mass at 21 μg to supermassive black holes at 10^7 M⊙. The evidence for dark matter (DM) is overwhelming, but arises purely from its gravitational interaction. The history of DM has passed through two phases. First as a binding force for galaxies. Second, the paradigm shift of n-body simulations to evolution of structure with dominant cold DM particles. Now comes the third phase, in which we zero in on distinctive predictions that arise from other interactions between baryons and DM. PBHs can help with problems besides DM, but all of these have alternative solutions.
Possible roles: DM, Microlenses, Triggering SNeIa, Binary inspirals, Hubble tension, Bulk flow tension, Early galaxy formation , JWST Little Red Dots, Pre-galactic globular clusters, Initiating comets.
Alternatives: WIMPs, Interstellar planets, No trigger required, Low S/N detections, Early dark energy, Rare event, Population III
Direct collapse black holes, Blue colour metallicity (not age) related, Tidal disruption
Earlier microlensing experiments were designed for ~solar masses with cadence ∼1 day. Roman permits a definitive search for Earth to asteroid mass PBH with cadence ∼minutes. The key questions are, what fraction of DM do PBHs comprise, and what’s the distribution over mass? We suggest supplementing the search underway by LSST with Rubin of ∼days, (peak rate of PBH discovery in the mass range M > 10^−4 M⊙) with Roman′s, extending it to 10^−9 < M < 10^4 M⊙. These are complementary, offering definitive detection of PBH as a component of DM. Moreover, detections would also constrain the mass function, connecting their formation to the conditions of the earliest moments of our Universe.
Co-Authors: Abi Saha, Jeremy Mould, Renee Key, Alan R. Duffy, R. Michael Rich & Ned Taylor
Transiting Planets around Ultracool Dwarfs with NASA’s Nancy Grace Roman Space Telescope
By: Philip Steven Muirhead
Type: Talk
Abstract: The continuous near-infrared monitoring of a dense stellar fields by the Nancy Grace Roman Space Telescope's Galactic Bulge Time Domain Survey provides an opportunity to discover transiting planets around mid-M and ultracool dwarf stars. I will present results from Tamburo, Muirhead & Dressing (2023), where we simulated the yield of small transiting exoplanets around the mid-M through T9 dwarfs that are expected to fall within the survey footprint. Using both local space density scaling and Galactic stellar population synthesis models, we predict the detection of over 1300 small transiting planets, including ~13 habitable-zone terrestrial worlds. These numbers are sensitive to planet occurrence rates around mid-M and ultracool dwarfs, which remain poorly constrained. Significant departures from our predictions would directly test whether small planet occurrence increases or decreases toward lower stellar masses relative to early-M dwarfs.
From JWST to Roman: A Unified View of Chemical Enrichment in Galaxies Enabled by Slitless Spectroscopy
By: Kalina Nedkova
Type: Talk
Abstract: Understanding star formation, feedback, and chemical enrichment across galaxy populations requires connecting low-metallicity systems to their massive, metal-rich counterparts. However, this connection has remained limited due to small sample sizes at both extremes. Now, deep JWST slitless spectroscopy programs are unlocking the low-mass, low-metallicity regime, while the upcoming Roman High-Latitude Wide-Area Survey (HLWAS) will provide the statistical power needed to characterize massive galaxies across cosmological volumes.
I will present recent results from the JWST/NIRISS PASSAGE program, which uses wide-field slitless spectroscopy that is free of photometric pre-selection to obtain spectra for all objects in the field of view. From a sample of >500 galaxies at 1.7 < z < 3.4, PASSAGE provides new constraints on the mass-metallicity relation down to 10^7 in stellar mass and metallicities as low as ~5% solar, extending measurements into a regime that has remained poorly sampled at intermediate redshift. These observations establish a critical foundation for interpreting future Roman observations.
The Roman HLWAS will obtain slitless spectroscopy over 2,400 square degrees, enabling metallicity measurement for >1 million high-mass, high-metallicity galaxies. Realizing this potential requires robust control of systematics unique to slitless spectroscopy, particularly emission-line misidentification, which can bias redshifts, metallicities, and inferred scaling relations. Using PASSAGE as a testbed, I will show that even with careful visual inspection, ~20% of single-line emitters are misidentified and I will discuss mitigation strategies based on multiple emission lines and photometric data.
Controlling for these systematics will be essential not only for galaxy evolution studies, but also for Roman’s cosmology goals, where systematic redshift errors directly affect measurements of large-scale structure and baryon acoustic oscillations. Together, deep JWST programs and Roman’s wide-area surveys will enable a unified view of galaxy assembly and chemical enrichment across cosmic time.
Utilizing the Power of Host Galaxies to Discover Transients with Roman
By: Anya Elizabeth Nugent
Type: Talk
Abstract: NASA’s Roman mission promises to illuminate the high-redshift transient sky, detecting many supernovae (SNe) and other transients out to z=3: currently uncharted territory for many transients. In turn, Roman will revolutionize our understanding of the evolution and deaths of stars across cosmic time. However, the majority of Roman-discovered transients will be incredibly faint and have scarce photometric coverage, making their classification and progenitor inference extremely difficult. Importantly, transient host galaxies provide a powerful and complementary avenue for distinguishing transients and understanding their origins, independent of transient emission. In this talk, I discuss how we can use transient hosts to identify the origins of high-redshift Roman transients. I will highlight recent efforts to rapidly infer transient host galaxy redshifts in Roman alerts. I will additionally discuss FrankenBlast - a novel tool that associates transients to their host galaxies and runs a machine-learning based SED-fitting algorithm to robustly constrain host stellar population properties, all within minutes per object. I present FrankenBlast-determined stellar population property results for ~10,000 host galaxies of YSE and ZTF-discovered transients. I highlight our key findings on how host properties can be used to identify events misclassified from traditional photometric classifiers and how host stellar population properties provide vital clues into transient progenitors. I conclude by outlining future plans to use tools like FrankenBlast for Roman data.
Spectroscopic Grism Simulation of Roman Touchstone Fields and Early-type Galaxies
By: Tathagata Pal
Type: Poster
Abstract: The Nancy Grace Roman Space Telescope (Roman) is going to revolutionize the way we investigate the Cosmos. Scheduled to launch no earlier than September 2026, Roman will provide us with petabytes of data whose science case includes exoplanet detection, cosmic expansion studies, as well as galaxy evolution studies. Realistic simulation work is needed to gauge the capabilities of Roman and make preparation for maximizing Roman’s science output. In this work, we have simulated Roman G150 grism scenes for a variety of stellar and galaxy targets. We simulate the Roman calibration touchstone fields (Euclid Deep Field North, Kepler field, and others) using catalogs from Gaia and synthetic stellar spectra from the BOSZ stellar spectral library. The results from these simulations can be used for various calibration purposes like modeling the effect of ice deposition on Roman’s Wide Field Instrument’s detectors. We also simulate the Roman High Latitude Wide Area Survey (HLWAS) footprint using catalogs from the OpenUniverse project and elliptical galaxy spectra simulated using CIGALE. These extragalactic simulations will enable us to use several spectroscopic features to investigate galaxy age and metallicity (e.g., using the CaT feature, Balmer and Fe absorption indices), sodium abundance (using the NaD feature), and Mg abundance (using the Mg2800 feature). The simulated spectra are fit using different stellar population synthesis models, and we investigate residuals between input and recovered stellar population parameters (like SFR, IMF exponent, and others). This study will also investigate the effect of contamination from other sources in Roman slitless spectroscopy.
Co-Authors: Dr. Gregory Mosby, NASA Goddard Space Flight Center, Greenbelt, MD, USA
Dust extinction variations unveiled by the Roman Galactic Plane Survey
By: Roberta Paladini
Type: Talk
Abstract: Variations of NIR dust extinction for lines of sight through the Galactic Plane have been reported in the literature. These variations appear to be significant, in particular, for dense molecular clouds in Star Forming regions, as well as towards the Galactic Center. One of the outstanding open questions is whether the NIR extinction curve in these environments is well represented by a single power law or it has a more complex behavior. The Roman Galactic Plane Survey will be performed by the Wide Field Instrument with an exquisite combination of sensitivity and angular resolution, therefore providing a unique large statistical sample enabling to set tight observational constraints and informing dust models with un-precedented accuracy.
Dark matter constraints from strongly lensed quasars - recent advances and perspectives for Roman
By: Hadrien Paugnat
Type: Talk
Abstract: The nature of dark matter (DM) remains one of the most fundamental open questions in cosmology, and the next frontier to test alternatives to the cold DM paradigm is the description of structure on sub-galactic scales. Strong gravitational lensing, as a purely gravitational probe, offers a unique window into this regime, enabling the detection of low-mass DM halos regardless of the presence of baryons, and at cosmological distances. In particular, flux-ratio anomalies in multiply imaged quasars ("quads") can reveal the presence of dark substructure, such that population-level statistics can be leveraged to constrain the particle nature of DM.
The Roman Space Telescope will deliver an unprecedented census of strongly lensed quasars, with the High Latitude Medium and Wide surveys expected to discover ~400 quads and ~4,000 doubly imaged quasars. Crucially, in addition to enabling lens discovery, the telescope will provide high-resolution imaging of quasar host galaxy arcs, needed to construct detailed lens models in the vicinity of the multiple images and break degeneracies between azimuthal lens structure and DM substructure. This should result in an order-of-magnitude improvement of DM constraints relative to current observational samples.
In this talk, we present recent advances and results from flux-ratio anomaly statistics, as well as perspectives for the Roman era. We will highlight: (i) competitive constraints on WDM set using warm dust flux ratios measured with the JWST/MIRI lensed quasar survey, (ii) narrow-line measurements with JWST/NIRSpec, paving the road for dual-source flux-ratio anomaly constraints, and (iii) a new approach to lens modeling using local lensing constraints, providing a conservative and independent check on the traditional macromodel-based inference. We will conclude with forecast dark matter results for Roman based on these new methods and results, which together define a robust and systematic-controlled analysis framework.
Co-Authors: Tommaso Treu (UCLA) Daniel Gilman (UChicago) Anna Nierenberg (UC Merced)
Sundial: Updated Simulations for the Roman Space Telescope's High-Latitude Time Domain Survey
By: Jillian Marie Paulin
Type: Poster
Abstract: The High-Latitude Time-Domain Survey (HLTDS) conducted by the Roman Space Telescope will search for Type Ia supernovae (SNe Ia) and other transient objects over the course of 5 years. We present Sundial, the most up-to-date set of HLTDS simulations based on current observing plans created by the Roman Time Allocation Committee (ROTAC). These simulations improve on the original 2025 Hourglass simulations by including all planned components of the survey (Pilot, Core, Extended) instead of just the Core, and by altering the observing strategy as plans change. Sundial contains simulated photometry and spectroscopy for SNe Ia, Core Collapse supernovae, and rare transients (tidal disruption events, kilonovae, pair-instability supernovae, and superluminous supernovae). Additionally, we use the SALT3 light-curve fitter to evaluate the quality of observations in preparation for future cosmological analyses, and demonstrate how these simulations are used in supernova classification with SCONE. These simulations are useful in a wide range of applications, including the calculation of volumetric rates of transients, calibration, training new classifiers, and more.
Creating Reddening Maps in M31 and M33 with Hubble and James Webb Space Telescope Data
By: Rachel Pauline
Type: Poster
Abstract: Celestial objects appear redder when dust preferentially absorbs blue light emitted from the object, referred to as reddening. This project focuses on creating reddening maps in two regions of the Andromeda galaxy using the red clump and red giant branch bump with data from the James Webb Space Telescope and the Hubble Space Telescope. These maps will help astronomers negate the effects of reddening to determine the true, unobscured characteristics of stars that they are observing in this galaxy. To make the maps, magnitude distributions were investigated in 8 bandpasses and a model comprised of a skewed gaussian, exponential function and two gaussian functions was applied to determine the best fit parameters for the data. Data that falls into the red clump and red giant branch bump was masked for one color combination using two Webb broadband filters and plotted on other color magnitude diagrams to display the shift in magnitude and color space between filters for each of the 7 colors. The steps in this project will also be applied to 2 regions in the Triangulum galaxy, and the results will be compared to other previously made reddening maps. These maps can also be used to compare the luminosity functions of Andromeda and Triangulum to those found in the solar neighborhood. Results of this project could also lead to determining what the tip of the red giant branch indicates for the James Webb Space Telescope.
Co-Authors: David Nataf
The Scientific Potential of Studies of Massive Quiescent Galaxies with Roman: A Precursor Catalog from Four Deep and Wide Extragalactic Fields
By: Sara Petty
Type: Poster
Abstract: Massive quiescent galaxies (MQGs) are important laboratories for a wide range of studies, ranging from the quenched endpoints of galaxy assembly to tracers of baryonic acoustic oscillations at intermediate redshifts. They will be primary targets for Roman in the HLWAS Medium, Deep, and Ultradeep tiers, where they can be efficiently selected over 0.5
Co-Authors: Duncan Farrah Mark Lacy Kristina Nyland
From HST Patchwork to Roman Panoramas: Mapping Globular Cluster Systems as Environmental Tracers in Dense Galaxy Clusters
By: Richard Thomas Pomeroy
Type: Poster
Abstract: Globular cluster systems (GCSs) provide long-lived tracers of galaxy assembly and environmental processing, but current studies of dense cluster environments are often limited by fragmented and heterogeneous imaging footprints. In our HST/ACS study of compact stellar systems (CSSs) in the Coma Cluster core, 26 individual pointings are required to cover a narrow, letterbox-like region spanning only ∼1 Mpc in R.A., even for Coma at a distance of ∼100 Mpc. We use these pointings to measure globular cluster (GC) and ultra-compact dwarf (UCD) populations, local backgrounds, and environmental trends around galaxies embedded in the dual-BCG environment of NGC 4874 and NGC 4889. However, combining these data into a coherent environmental analysis requires careful treatment of heterogeneous depth, incomplete spatial coverage, edge effects, local-background estimation, and completeness corrections. Despite these limitations, the HST data reveal evidence that GCSs in Coma retain measurable signatures of environmental processing. In particular, the metal-poor/blue GC population appears especially sensitive to the dense cluster-core environment: blue-GC fractions, relative to the cluster mean, correlate more strongly with local CSS density and BCG-related tidal structure than with simple cluster-centric radius. These results suggest that GC redistribution is governed by a multi-scale environment, from galaxy-scale halos to the extended influence of the dual BCGs. Roman’s combination of high angular resolution and wide-field near-infrared imaging can transform this type of analysis. Contiguous Roman imaging of nearby galaxy clusters would allow GCs, UCDs, and intracluster compact stellar populations to be mapped across cluster-scale environments with far less dependence on stitched, inhomogeneous datasets. I will discuss how Roman-era observations can turn current HST-based techniques into scalable environmental diagnostics, enabling cleaner tests of GC stripping, redistribution into intracluster space, and the connection between galaxy halos and cluster assembly.
Co-Authors: Juan P. Madrid
Extragalactic Star Clusters and Dwarf Galaxies with Roman
By: Aaron Romanowsky
Type: Poster
Abstract: The smallest and faintest stellar systems -- star clusters and dwarf galaxies -- provide unique constraints on star and galaxy formation and on the nature of dark matter itself.The depth, resolution, field of view, and near-infrared wavelengths of Roman promise to open up transformative studies of these objects in the nearby Universe, through discovery and through modeling of stellar populations. I will present plans for a systematic survey of thousands of low-redshift globular clusters in the core surveys, in synergy with other surveys such as Rubin/LSST, and interpreted through models for galaxy formation. I will also showcase results from Euclid on nearby star clusters and dwarfs as a preview of the potential of Roman.
Co-Authors: Yimeng Tang, Oleg Gnedin, Zachary Coustier, David Li, Yasna Ordenes-Briceno, Kevin Bundy
The contribution of supernovae to the cosmic dust budget
By: Sam Rose
Type: Talk
Abstract: Despite the ubiquitous presence of dust across cosmic time, the formation of dust remains an open question. Large amounts of dust can form in the ejecta of supernovae, but supernova shocks may also destroy large quantities of dust. Disentangling newly forming dust in the ejecta from pre-existing dust in the CSM also complicates measurements of supernovae dust production. Therefore, the overall contribution of supernovae dust to the cosmic dust budget is debated. In this talk, I will present results from the Red Astronomical Transient (RAT) survey, which probes the early signatures of new dust formation and tests the presence of pre-existing CSM dust by combining data from the Zwicky Transient Facility (ZTF) alert stream with ground-based optical and NIR follow-up observations with WINTER and Keck II/NIRES. With the large sample of supernovae discovered and characterized by ZTF it is possible to obtain a larger and more complete sample of dust and CSM diagnostics in core-collapse supernovae including narrow line ejecta-CSM interaction features, NIR excesses, and the spectroscopic detection of molecular precursors to dust formation. I will highlight the upcoming role of Roman which will provide sensitive NIR observations that are vital to constrain supernova dust production at late times.
Co-Authors: Mansi Kasliwal, Ryan Lau, Jacob Jencson, and the ZTF and WINTER teams
Investigating Weak Lensing Effects on Supernovae Cosmology
By: Benjamin Rose
Type: Talk
Abstract: The use of Type Ia supernovae (SNIa) as standard candles allows researchers to estimate cosmological distances precisely due to their low level of intrinsic scatter and standard luminosity. At high redshifts, the effects of weak gravitational lensing dramatically increase. Up until now, SNIa have been measured up to redshifts of 1 with the lensing effect being approximated by the linear model of (0.055 +/- 0.04)z. With the Nancy Grace Roman Space Telescope, will be able to be measured to redshifts larger than 3. This future increase in the range of redshifts capable of being measured with Roman has brought forward the question of if the current linear model for weak lensing effects will continue to accurately measure the scattering effects caused by weak lensing at high redshifts. I will present research that used DC2 galaxies and DES weak lensing to simulate a Roman supernova cosmological sample allowing us to estimate and test ways to measure weak lensing effects on SNIa cosmology and investigate initial measurements of sigma_8 with SNIa.
Co-Authors: Jenna Glaze, Jansyn Pihl
Mapping Dark Matter from the Stratosphere: A First-of-Its-Kind Weak-Lensing Mission (SuperBIT)
By: Sayan Saha
Type: Poster
Abstract: Merging galaxy clusters provide a unique laboratory to study the interplay between baryonic and dark matter. X-ray emission, the SZ signal, and galaxy richness trace the baryonic component, while weak gravitational lensing offers the most direct probe of the total, predominantly dark matter distribution, requiring highly precise galaxy-shape measurements.
I will present results from the Super Pressure Balloon-borne Imaging Telescope (SuperBIT), a first-of-its-kind stratospheric weak-lensing mission targeting ~30 merging galaxy clusters. SuperBIT paves the way for space-quality imaging at orders of magnitude lower cost than traditional space telescopes, offering a promising model for future weak-lensing surveys. I will briefly describe the galaxy-shape measurement pipeline we developed (arXiv:2603.18376), highlight key systematics, and show preliminary convergence maps of several clusters — the first weak-lensing mass reconstructions from the stratosphere.
As a balloon-borne observatory, several of the challenges in weak-lensing shape measurement are closely analogous to those faced by space-based missions such as Roman — for example, PSF undersampling and high-frequency structure in the PSF. I will discuss these challenges and the solutions we adopted, and briefly address how SuperBIT's blue-band data adds complementarity to Stage IV surveys' red and near-infrared imaging.
Co-Authors: Jacqueline E. McCleary, Spencer W. Everett, Maya Amit, Georgios N. Vassilakis, Emaad Paracha, Leo W.H. Fung, Steven J. Benton, William C. Jones, Gavin Leroy, Eric M. Huff, Richard Massey, Thuy Vy T. Luu, Ajay S. Gill, Mohamed M. Shaaban, Philippe Voyer, Anthony M. Brown, Giulia Cerini, Paul Clark, Matthew Craigie, Christopher J. Damaren, Tim Eifler, David Harvey, Eric Habjan, John W. Hartley, Bradley Holder, Mathilde Jauzac, David Lagattuta, Jason S.-Y. Leung, Lun Li, Johanna M. Nagy, C. Barth Netterfield, Susan F. Redmond, Jason D. Rhodes, Andrew Robertson, L. Javier Romualdez, Jurgen Schmoll, Ellen Sirks, Sut Ieng Tam, André Z. Vitorelli, Alfredo Zenteno
Evolved supergiants in the Local Volume: Supernova progenitors, feedback, mass-loss and more!
By: Sumit Sarbadhicary
Type: Talk
Abstract: Evolved counterparts of massive OB stars, such as red and yellow supergiants, are vital for understanding multiple areas of astrophysics, from the progenitors of supernovae, to supernova feedback that drive galaxy evolution, models of binary stellar evolution, and much more. Roman’s wide field-of-view in the near-infrared provides an unprecedented opportunity to map the population of these cool supergiants across the entire star-forming disks of Local Volume galaxies, concurrently serving as a much-needed permanent reference for future supernova progenitors. In my talk, I will discuss results of existing searches in the inner disks of ~80 nearby (<20 Mpc) galaxies with Hubble and JWST as part of the Physics at High Angular Resolution in Nearby Galaxies (PHANGS) and Local Group L-Band Surveys (LGLBS), with recoveries of nearly 100,000 supergiants. I will discuss novel results from comparison of these catalogs with existing ISM data, including: 1) atomic and molecular ISM data from VLA and ALMA, showing that a significant fraction of massive stars explode outside molecular clouds, supporting the necessity of more realistic stellar feedback models in galaxy simulations that include stellar winds and photoionization (in addition to supernovae), and 2) high-resolution dust maps from JWST MIRI, revealing massive evolved supergiants associated with circumstellar dust, and how they inform pre-supernova mass-loss models in cool supergiants. Finally, I will discuss analysis strategies using Roman HLWAS and GAS observations on extending catalogs of such supergiants all the way to the outer UV-bright star-forming disks of Local Volume galaxies, whose large angular extents have thus far been outside the reach of Hubble and JWST's fields of view.
Co-Authors: Adam Leroy (OSU), David Thilker (JHU), Erik Rosolowsky (U Alberta), Janice Lee (STSCI), Eric Koch (CfA), PHANGS + LGLBS collaborations
Small Galaxies, Big Questions: Dwarf Galaxies Through Optical Spectroscopy
By: Zahra Sattari
Type: Poster
Abstract: Low-mass galaxies provide a sensitive probe of the physical processes that regulate galaxy growth, including star formation, dust attenuation, feedback, and gas cycling. Because of their shallow potential wells and relatively low dust content, these systems are expected to respond strongly to both internal feedback and external environmental effects. I will present a spectroscopic study of low-mass galaxies in the COSMOS field, focusing on their star formation rates, dust attenuation properties, and the connection between nebular and stellar reddening. Using optical spectroscopy, we find that low-mass galaxies follow a nearly scale-free SFR–stellar mass relation. At the same time, their dust attenuation behavior differs from that of more massive galaxies, with lower and less mass-dependent nebular attenuation, suggesting that feedback and efficient dispersal of birth clouds play an important role in shaping their interstellar medium. I will also discuss how this work connects to ongoing JWST/PASSAGE studies of low-mass galaxies in and around a protocluster at cosmic noon. By extending measurements of the SFR–mass relation and gas-phase metallicity to lower stellar masses in both field and overdense environments, these observations provide a path toward testing whether environment affects the baryon cycle during the epoch of peak galaxy assembly. Together, these studies use low-mass galaxies as laboratories for understanding how star formation, feedback, dust, and environment jointly shape galaxy evolution across cosmic time.
Co-Authors: Vihang Mehta, Daniel Kelson, Nima Chartab, Harry Teplitz, Bahram Mobasher, Shannon Patel and the PASSAGE team
The Lumina Simulation & Puzzles of high-redshift galaxies in the Roman era
By: Xuejian Shen
Type: Talk
Abstract: JWST has revealed a far more vibrant high-redshift Universe than the canonical galaxy formation model predicts, with an overabundance of massive and bright galaxies, alongside the early emergence of morphologically mature galaxies, quenched galaxies, and massive protoclusters. This growing body of evidence motivates us to consider a class of beyond-ΛCDM models that enhance the primordial power spectrum on small scales not yet constrained by traditional observational approaches. In this talk, I will discuss several examples, including early dark energy, axion-like dark matter, and primordial features from inflation, each typically motivated by independent physics considerations. As a case study, I will present promising results from a new large-volume cosmological radiation-hydrodynamical simulation of galaxy formation at z>6 in early dark energy: the largest cosmological hydrodynamical simulation, by number of resolution elements, ever achieved for an alternative cosmological model. This simulation predicts an accelerated phase of structure formation in the high-redshift Universe, boosting dark matter halo and galaxy abundance and reproducing many JWST observations with a standard galaxy formation model. Crucially, this scenario predicts distinct signatures in galaxy clustering, including a modest shift in the baryon acoustic oscillation peaks and a reduction in galaxy and halo bias. Other models can imprint different clustering features, and the upcoming wide-field surveys of the Roman Space Telescope are uniquely positioned to measure them. Roman will deliver an independent, statistically powerful test of these models, one with strong synergies with JWST. I will close by presenting predictions for Roman from our simulations.
Co-Authors: Oliver Zier, Mark Vogelsberger, Michael Boylan-Kolchin, Lars Hernquist, Sandro Tacchella, Rohan P. Naidu
Hot Jupiter Occurrence Rate and Atmospheric Characterization Beyond the Solar Neighborhood
By: Avi Shporer
Type: Talk
Abstract: We present our plan to measure the occurrence rate and atmospheric properties of transiting hot Jupiters (HJs), by targeting HJs with detectable thermal emission in the Roman GBTDS. The detection of thermal emission - in the secondary eclipse and thermal emission phase modulation - allows to differ between real planets and false positive transiting planet candidates, that is otherwise not possible as the targets are too faint for traditional techniques. The F146 near-IR sensitivity enables detection of secondary eclipses and thermal phase curve modulations for HJs orbiting main-sequence stars at 7–8 kpc — over an order of magnitude beyond current surveys which are confined almost entirely to within 1 kpc. Our planet yield estimate, based on Besançon Galaxy Model stellar catalogs for all 6 GBTDS fields, predicts 858-1,232 HJs with both thermal phase curves and secondary eclipses detectable at >10σ, using two independent occurrence rate estimates. This work addresses two science goals. First, we will measure the HJ occurrence rate in the metal-rich Galactic bulge, testing whether the steep local metallicity–occurrence correlation extends to Galactic scales. Second, for each detected system we will extract atmospheric parameters, providing the first statistical census of HJ atmospheres beyond the solar neighborhood.
Co-Authors: Ian Wong, STScI
Spectroscopic forecast of z>7 quasar population with Roman
By: Mainak Singha
Type: Talk
Abstract: JWST suggests that AGN could provide a 31-75% of the ionizing photon budget at z=4.5-6.5, but this conclusion hinges on faint Type I and Type II AGN populations whose abundance beyond z>7 remains essentially unconstrained. The critical missing measurement is therefore simple: can wide-area slitless spectroscopy find the faint accreting black holes that would determine whether AGN still contribute meaningfully during the bulk of reionization? Roman's WFI grism is designed for exactly this measurement: 1.0-1.93 micron slitless spectroscopy over thousands of square degrees, covering the redshifted Lyman-alpha break of z>7 quasars. We forecast Roman's reach at z=7.5-10 under HLWAS-like exposures using the ESpRESSO simulator (Gabrielpillai et al. 2026), spanning three Lyman-alpha equivalent-width priors and testing both break detectability and spectral-shape classification against M/L/T dwarfs and red galaxies. Roman recovers z>7.5 quasars to J_AB ~24.5, extending to ~25 in the highest-equivalent-width prior. Among detected sources, the classifier achieves TPR >= 0.982 with zero false positives across ~19,000 realizations, corresponding to 95% upper limits of <3.7e-4 for dwarfs and <2.6e-4 for red galaxies. At z~8, J_AB ~24.5-25 maps to L_bol ~1e45-1e46 erg/s, overlapping the faint Type I regime that dominates AGN ionizing emissivity in current z<7 models. Roman would therefore provide the first wide-area grism constraints on the faint z>7 AGN luminosity function, directly testing whether the AGN contribution inferred at z=4.5-6.5 persists into the reionization era.
Co-Authors: Sangeeta Malhotra, James Rhoads and Isak Wold
Reionization in the Roman Era: Developing a Flexible Simulation Framework for Multi-tracer Cross-Correlations
By: Guochao Sun
Type: Talk
Abstract: Roman will soon transform reionization science by mapping enormous volumes of the high-redshift Universe, but fully exploiting this opportunity requires simulations that connect Roman-selected galaxies to diverse observables of the ionization, thermal, and velocity structure of the intergalactic medium. I will present progress on our ongoing Roman wide-field science program to develop a flexible simulation framework for Roman-era reionization cross-correlation science, including both an overview of the modeling architecture and a concrete worked example. I will first describe recent extensions of the LIMFAST semi-numerical code to generate, from large-volume N-body simulations, self-consistent realizations of Lyman-break galaxies, Lyman-alpha emitters, intensity maps of HI 21 cm and other emission lines, and the patchy kSZ signal. By combining large-volume structure formation, halo-based galaxy modeling, ionization and thermal evolution of the IGM, Lyman-alpha transmission, and realistic survey selection in a common framework, these simulations allow us to identify which cross-correlations are most robust to foregrounds and systematics, which astrophysical and cosmological parameters they constrain, and how Roman can best join forces with line intensity mapping and CMB surveys. I will then walk through a case study on Roman photometric galaxies and foreground-filtered 21 cm maps. Standard two-point correlations are strongly limited during the EoR because photometric redshift errors erase small-scale line-of-sight galaxy modes, while 21 cm foreground mitigation removes the large-scale modes needed for direct Fourier-space overlap. We show that cross-correlating galaxies with the filtered-then-squared 21 cm field restores sensitivity through higher-order mode coupling, allowing Roman photometric samples to recover information that would otherwise appear inaccessible. Forecasts for Roman and SKA-Low surveys indicate that this signal should be detectable across much of reionization and can constrain key aspects of early galaxy formation and evolution, including the escape fraction, X-ray heating, and the mass dependence of star formation efficiency.
Co-Authors: Adam Lidz (Penn), Tzu-Ching Chang (Caltech/JPL), Jordan Mirocha (Caltech)
Star Formation in Low-Metallicity Environments
By: Grace Telford
Type: Talk
Abstract: Energetic feedback from low-metallicity massive stars regulates the evolution of low-mass galaxies, both nearby and in the early Universe. To understand those processes requires robust models of massive star formation, evolution, and feedback at low metallicity. Yet, these models remain largely theoretical and uncertain due to a lack of observational constraints in metal-poor galaxies close enough that individual stars and star-forming gas can be spatially resolved. I will present the recent discovery of molecular gas in a star-forming galaxy at just 3% Solar metallicity with JWST, which provides a novel constraint on the nature of the extremely metal-poor gas from which stars formed in the early Universe. I will then discuss recent insights into the astrophysics of massive stars in nearby dwarf galaxies and the implications of their observed properties for stellar feedback models at low metallicity. Finally, I will highlight opportunities to study massive-star formation and feedback in local dwarf galaxies with Roman.
Robustness of Reionization Constraints from Lyα Emitters in the Roman Era
By: Jahaan Thakkar
Type: Poster
Abstract: Lyα-emitting galaxies provide one of the most direct routes for mapping the ionization state of the intergalactic medium during the epoch of reionization, and they will be a key high-redshift population for wide-field near-infrared surveys with the Nancy Grace Roman Space Telescope. I present a forward-modelling framework for testing how robustly Lyα luminosity functions can be converted into constraints on the global ionized fraction. Dark matter halos from 21cmFAST reionization simulations are populated with galaxies using empirical UV–halo abundance matching, assigned intrinsic Lyα luminosities through rest-frame equivalent-width prescriptions, and propagated through inhomogeneous IGM fields using sightline-by-sightline Voigt-profile optical-depth calculations. Bayesian comparison with observed Lyα luminosity functions then yields the redshift evolution of the volume-averaged ionized fraction. The central result is that the inferred reionization history is strongly sensitive to assumptions about intrinsic Lyα emission. Replacing a commonly used exponential equivalent-width distribution with a JWST-calibrated log-normal distribution produces substantially different Lyα luminosity functions and mutually inconsistent ionization histories, even when the IGM fields, halo population, UV continuum model, and radiative-transfer treatment are held fixed. By contrast, changes to the emergent Lyα line-profile shape, such as adopting a skewed log-normal profile rather than a Gaussian, introduce only moderate shifts in the inferred ionized fraction and remain subdominant to the equivalent-width prescription. These comparisons identify the intrinsic equivalent-width distribution as the dominant systematic in current LAE-based reionization inference. This result is directly relevant for Roman-era high-redshift galaxy science. Roman’s wide-field imaging and slitless spectroscopy will greatly expand LAE samples across rare environments and large cosmic volumes, reducing statistical uncertainties and cosmic variance. Exploiting that statistical power will require forward models whose galaxy-scale Lyα prescriptions are calibrated as carefully as their IGM physics. I argue that robust Roman constraints on reionization will depend on jointly modelling LAE selection, intrinsic equivalent-width distributions, and IGM transmission.
Co-Authors: Soumak Maitra, Girish Kulkarni
Modeling non-linear clustering in Roman-Line intensity mapping cross-correlations
By: Rakshitha Thaman
Type: Talk
Abstract: Roman is anticipated to bring transformative insights to the study of large scale structure (LSS) and cosmology with its superb capabilities and unprecedented field of view. One approach to maximize Roman’s scientific return is to harness the synergy with line intensity mapping (LIM), which accounts for emission from all galaxies along the line of sight across hundreds of square degrees. Cross-correlations between [C II] line-intensity maps and wide-field galaxy surveys such as Roman provide a powerful way to constrain both galaxy evolution and cosmology across cosmic time. With the deeper observations expected from Roman, these measurements will probe clustering into the non-linear regime. Fully exploiting the information content of upcoming surveys therefore requires accurate modeling of the LIM - galaxy cross-power spectrum, including non-linear bias terms and cross-shot noise.
To construct such a model, we use the Santa Cruz semi-analytic model to generate physically motivated mock galaxy catalogs and [C II] intensity maps from a 2 deg² lightcone. We then use a non-linear model based on the Effective Field Theory of Large-Scale Structure (EFTofLSS), previously tested on mock galaxy and LIM auto-power spectra but not yet on the cross-power spectrum. Using MCMC methods, we fit this model to the mock cross-power spectrum and infer posterior constraints on the galaxy bias, intensity-weighted [C II] bias, and cross-shot noise parameters. Since [C II] emission is known to be a tracer of star formation, these measurements can also be used to constrain the cosmic star formation rate over time.
Co-Authors: Anthony Pullen, L.Y. Aaron Yung, Azadeh Moradinezhad Dizgah
Illuminating Dark Energy and Black Holes with Strong Gravitational Lensing in the Nancy Grace Roman Space Observatory Era
By: Kim-Vy Tran
Type: Talk
Abstract: With recent and upcoming all-sky surveys such as Euclid, LSST, and Roman, deep high resolution imaging of increasingly vast cosmological volumes is now available. Using machine learning to search for the distinct visual signature of strong gravitational lensing, the AGEL survey identified thousands of strong lensing candidates that span a wide range in mass. The AGEL lenses include deflectors at z>0.5 that are ideal for follow-up spectroscopic studies to address a range of questions in astrophysics and cosmology. I highlight AGEL results including cosmography with compound lenses, evolution of mass density profiles, and our discovery of Einstein spirals that can be used to weigh quiescent Super-Massive Black Holes to z~1.
We plan to mine the Roman High Latitude Wide-Area Survey to identify ~20 compound lenses and ~20 Einstein spirals to (i) deliver a benchmark measurement of the dark energy equation of state parameter 𝑤 and (ii) cement a new probe for SMBH–host galaxy co-evolution from 0.3 ≲ 𝑧 ≲ 1.
Co-Authors: Rodrigo Cordova Rosado, Tania Barone, Giovanni Ferrami, Nandini Sahu, Keerthi Vasan, Duncan Bowden, Tucker Jones, Karl Glazebrook, Anowar Shajib, and the AGEL team
The Roman Coronagraph Instrument: Pathway to Imaging Reflected Light Planets and Habitable Worlds
By: Jason Wang
Type: Talk
Abstract: The Nancy Grace Roman Space Telescope Coronagraph Instrument is a new space-based visible-wavelength high-contrast imager and spectrograph. It is expected to achieve 100-1000x better contrast ratios than current high-contrast imaging systems and has the potential to image the first planets in reflected starlight (as well as their circumstellar environments). This significant jump in sensitivity is made possible because the Roman Coronagraph will combine deformable mirrors, photon-counting electron-multiplying CCDs, high-performance coronagraph masks, and wavefront sensing and control on a space telescope for the first time. Demonstrating how these technologies work together is a critical step towards imaging Earth analogues with the Habitable Worlds Observatory. This talk will provide an overview of the instrument, anticipated performance, instrument goals, planned science objectives, and initial observing plans.
The Environment and Clustering of Distant Quasars and AGNs in the JWST+Roman Era
By: Feige Wang
Type: Talk
Abstract: The environments of distant quasars and active galactic nuclei (AGNs) provide a unique window into the formation of the earliest supermassive black holes, their host galaxies, the dark matter halos they inhabit, and the emergence of large-scale structure within the first billion years of cosmic history. Over the past two decades, HST observations have revealed mixed evidence for galaxy overdensities around high-redshift quasars, limited by small fields of view, projection effects, and incomplete spectroscopy. More recently, JWST has transformed this field through deep NIRCam WFSS and NIRSpec spectroscopy, uncovering significant galaxy overdensities around some z>6 quasars while also revealing substantial diversity in their environments. In particular, the JWST ASPIRE project has provided robust clustering measurements of galaxies around quasars in the reionization era, revealing that luminous reionization-era quasars reside in massive dark matter halos and serve as effective signposts of protoclusters. However, current studies remain limited to relatively small samples and narrow survey areas, and therefore still suffer from large uncertainties caused by cosmic variance.
The Roman Core Community Surveys, together with future deeper General Astrophysics Survey (GAS) observations around quasars and AGNs, will open a fundamentally new regime for studying the environments of distant quasars, AGNs, and little red dots. Roman’s combination of HST-like spatial resolution and an ultra-wide field of view will enable efficient mapping of megaparsec-scale structures surrounding these biased tracers across statistically significant samples. Roman surveys will bridge the gap between the detailed but pencil-beam JWST observations and wide-area studies of cosmological structure formation, enabling systematic measurements of AGN halo masses, environmental diversity, and the connection between black hole growth and large-scale structure during cosmic dawn. In this talk, I will first review the lessons learned from recent JWST observations and then discuss future opportunities enabled by Roman.
Co-Authors: The EREBUS and COSMOS-3D collaborations
Asteroseismology with the Galactic Bulge Time Domain Survey
By: Trevor Weiss
Type: Talk
Abstract: Roman’s Galactic Bulge Time-Domain Survey presents a unique opportunity for asteroseismology in the Galactic bulge, promising to revolutionize our understanding of the formation history of the bulge and of the properties of its present-day stellar populations. We have previously shown that Roman’s cadence, depth, and infrared sensitivity enable detections of global asteroseismic parameters, including 𝞶max and Δ𝞶, for hundreds of thousands of red clump and red giant branch stars. Here, we present revised simulations that include updated survey strategies, a revision of asteroseismic detection probability, and luminous stellar populations. The latter are expected to contribute 10% of the total expected asteroseismic yield, and offer a view into distinct stellar populations for understanding the kinematics and evolution of the bulge. We will also discuss ongoing preparatory work for asteroseismology with Roman, including saturated photometry and assembling a red giant target list.
Co-Authors: The Roman Asteroseismology Wide Field Science Team
Transits in the Roman galactic EXoplanet Survey (TRExS): Updated Exoplanet Yield Estimates and Early Science Opportunities
By: Robert F Wilson
Type: Talk
Abstract: Pixel-level simulations have predicted that the Galactic Bulge Time Domain Survey (GBTDS) will facilitate the discovery of ~100,000 hot and warm transiting exoplanets, ranging in size from Earth to Jupiter. This nearly order of magnitude increase in the known transiting exoplanet sample will enable powerful demographic studies, probe diverse Galactic populations by leveraging Roman's sensitivity to faint stars, enable the detection of thousands of exoplanet atmospheres due to the near-infrared observations where planet–star contrast ratios are favorable, and form a complete picture of hot and cold giant exoplanet populations when combined with the microlensing detections. This talk will present transiting planet yield estimates for the final GBTDS survey design, offer opportunities for early transiting exoplanet science that can be accomplished within the first two years of the GBTDS, and advocate for long-term science goals that are poised to revolutionize multiple exoplanet subfields.
Co-Authors: Elisa Quintana, Jorge Martinez-Palomera, Eamonn Kerins, Kelsey Hoffman, Nestor Espinoza, Susan Mullally, Alison Duck
A Prominent Galaxy Overdensity Candidate within the First 500 Myr
By: Zihao Wu
Type: Talk
Abstract: We report a candidate galaxy overdensity at z~10.5 identified in the JWST Advanced Deep Extragalactic Survey (JADES), offering a rare view of large-scale structure formation within the first 500 Myr of cosmic history. The overdensity contains 18 galaxies with consistent photometric redshifts within 8 comoving Mpc in projection. Its galaxy number density is four times higher than the field expectation, accounting for one-third of comparably bright galaxies and nearly 50% of the total star formation rate at 10 < z_phot < 12 in the GOODS-S field. Galaxies in the overdensity more frequently exhibit close companions or substructure than field galaxies, with one-third showing such features within 1 kpc at consistent photometric redshifts, suggesting enhanced interactions in a dense early environment. Most members have stellar masses of 0.6-3 x 10^8 Msun, half-light radii of ~200 pc, and star formation rates of ~5 Msun yr^-1. Their properties are modestly elevated relative to field galaxies but remain broadly consistent with established high-redshift scaling relations. We find tentative evidence for spatially varying Ly-alpha transmission inferred photometrically, consistent with the emergence of an ionized bubble. I will also discuss our upcoming JWST spectroscopic follow-up program targeting this system to study its Ly-alpha emission and local reionization environment. This overdensity highlights the growing ability of JWST to uncover rare overdense environments at cosmic dawn and illustrates the discovery space that future wide-field surveys with the Nancy Grace Roman Space Telescope will open for mapping the earliest large-scale structures and their role in cosmic reionization.
Co-Authors: Daniel J. Eisenstein, Benjamin D. Johnson, Kevin Hainline, William M. Baker, Andrew J. Bunker, Alex J. Cameron, Emma Curtis-Lake, A. Lola Danhaive, Ryan Hausen, Jakob M. Helton, Zhiyuan Ji, Tobias J. Looser, Roberto Maiolino, Petra Mengistu, Pierluigi Rinaldi, Brant E. Robertson, Fengwu Sun, Sandro Tacchella, James A. A. Trussler, Christina C. Williams, Christopher N. A. Willmer, Joris Witstok
Optimal Summary Statistics for Constraining Reionization Morphology with LAEs Using Roman
By: Michael Wyatt
Type: Talk
Abstract: Reionization marks the epoch when light from the first stars, galaxies, and active galactic nuclei shone outwards, ionizing the once neutral hydrogen which comprised the intergalactic medium (IGM). These early luminous sources left their mark in the form of enormous ionized bubbles surrounding them, connecting the Universe across an equally enormous range of scales. This connection makes reionization a unique and promising probe of the early Universe, as observable, large-scale signatures can help to elucidate the small-scale details of galaxy formation and evolution. In particular, models and simulations have revealed that the morphology of ionized bubbles (their sizes and distributions) are sensitive to a variety of galaxy parameters. One of the most promising methods of constraining the bubble morphology is through the observation of Lyman-alpha emitting galaxies (LAEs), whose observed distribution is modulated by the morphology of ionized bubbles due to the enormous Lyman-alpha optical depth of the neutral (but not ionized) IGM. There exist a multitude of summary statistics for quantifying the distribution of LAEs. In this work, we compare popular existing statistics such as the angular correlation function and void statistics, as well as introduce a new statistic taking advantage of persistent homology for constraining the morphology of ionized bubbles. The Roman Space Telescope will offer an unprecedented set of observations of LAEs, with a combination of both wide-field surveys and spectroscopic redshifts. With this in mind, we specifically consider observational strategies made possible by Roman, and provide insight into the most promising summary statistics for upcoming data.
Co-Authors: Steven R. Furlanetto, Andrei Mesinger, Masami Ouchi, Hiroya Umeda, Nikolaos Triantafyllou, Samuel Gagnon-Hartman
Developing catalog-based searches for dwarf galaxies in stellar halos with Roman
By: Kathryn Wynn
Type: Poster
Abstract: The upcoming launch and commissioning of the Nancy Grace Roman Space Telescope (Roman) will advance our knowledge in the fields of galactic structure and evolution faster than ever before. With its Wide-Field Instrument, it will be able to observe large areas of the stellar halos of nearby galaxies in hours, rather than months. To prepare for this massive influx of data, we are developing a pipeline that uses a catalog-based approach to quickly produce and probe recovery rates of simulated dwarf galaxies within larger stellar halos. While this approach is less detailed than full end-to-end image simulations, it provides a much faster way to explore a large parameter space. Using a catalog-based approach allows us to predict idealized, high-level recovery rates of dwarf galaxies as a function of several different user-input parameters (such as mass, age, metallicity, etc.). Similarly, using an event-based pipeline allows for any step in the process to be repeated or restarted easily if the user wishes to change a parameter or finds an error. The ultimate goal of this pipeline development is to provide a tool that quickly and easily provides a ‘first-look’ at recovery rates in a large parameter space. This will help to determine the observational strategy required to resolve structures in extended stellar halos of nearby galaxies, which are sensitive to dark matter halo structure and galaxy accretion histories. This work is part of the RINGS program funded by NASA under grant 80NSSC24K0084.
Co-Authors: Ben Williams, Eric Bell, RINGS team
Roman hIgh-redshift transient SciencE (RISE): Enabling a Non-Cosmology Time-Domain Program
By: Lin Yan
Type: Talk
Abstract: High-redshift supernovae (SNe) are emerging as key probes of stellar evolution, galaxy environments, and fundamental physics in the early universe, with recent JWST discoveries hinting at evolving SN properties, rates, and progenitors at z>2. Progress is currently limited by small, heterogeneous samples and sparse temporal coverage, motivating the need for a large, uniform, well-cadenced, and spectroscopically confirmed high-redshift transient sample. We present the Roman hIgh-redshift transient SciencE (RISE) program, a new Roman WFI initiative designed to deliver to the community deep stacked imaging, image subtraction, and faint alert streams that extend beyond single-epoch detections. RISE is an augmentation of the RAPID project, pushing sensitivity deeper by 1 - 2 magnitudes, enabling the detections and detailed studies of high-redshift transients. It will also provide post-alert characterization and prioritization tools, including forced-photometry light curves, host-galaxy photometric redshifts, color and light-curve constraints, and machine-learning-based classification, enabling efficient identification of the most compelling high-z transients for follow-up. In combination with JWST and Rubin Observatory observations, RISE will unlock a statistically robust view of transient populations across cosmic time and open a new frontier in high-redshift time-domain astrophysics.
Co-Authors: O. Fox, M. Kasliwal, R. Rest, B. Rusholme, J. Jencson, S. van Dyk, T. Moriya, M. Siebert, D. Coulter, Y. Yao, Strogler, R. Quimby, R. Sterin, J. Pierel, M. Li, J. Santana, R. Lau, M. Shahbandeh
Probing early supermassive black holes using Roman and LSST
By: Jinyi Yang
Type: Talk
Abstract: Observations in the nearby Universe have established close connections between supermassive black holes (SMBHs) and the properties of their host galaxies, suggesting that SMBHs play an important role in galaxy evolution. Yet the assembly history of SMBHs and the evolution of the BH–host relation, particularly in the early Universe, remain poorly understood. Recent observations of z>6 quasars/AGN have raised major questions about early SMBH formation, BH growth, and BH-host co-evolution, many of which remain unresolved. Now, thanks to the capabilities of new cutting-edge facilities, we are entering a new era of investigating these SMBHs. Leveraging the upcoming Roman deep imaging and spectroscopic data, in synergy with LSST, we will carry out a survey to cover a broad dynamic range of SMBHs over cosmic epochs, from the high-mass end in luminous quasars to the low-mass end in faint AGNs, as well as "little red dots" (LRDs), a mystery population recently discovered with JWST, suggesting a significant fraction of SMBH growth remained undetected by traditional surveys. Specifically, in the first two years, the dataset allows us to systematically search for z>7-8 quasars/AGNs, z∼1 LRDs, and reddened or obscured AGNs across redshifts. These efforts will pave the way for SMBH studies using the full HLWAS field, and the resulting samples will trace SMBHs across different growth phases and cosmic epochs, enabling a comprehensive view of SMBH demographics.
Weak Baryonic Suppression in Astrid: A Challenge for Roman Weak Lensing Cosmology
By: Yanhui Yang
Type: Talk
Abstract: Upcoming weak lensing surveys with the Nancy Grace Roman Space Telescope will play an important role in probing tensions in cosmology such as the S_8 discrepancy between early- and late-Universe measurements of matter clustering. Interpreting Roman observations will require accurate theoretical predictions for the nonlinear matter power spectrum, including robust modeling of baryonic effects. We present new results from the Astrid cosmological hydrodynamical simulation and its dark matter-only counterpart, Astrid-DMO, focusing on the impact of active galactic nucleus (AGN) feedback on the large-scale structure. We find substantially weaker suppression of the matter power spectrum than in many existing state-of-the-art simulations, with no significant suppression at z = 0 and only mild suppression at intermediate redshifts. Through controlled small-volume simulations, we identify a key driver of this discrepancy as the treatment of black hole (BH) dynamics. Widely used BH repositioning schemes artificially enhance BH mergers and boost kinetic AGN feedback, leading to stronger suppression of matter clustering. In contrast, Astrid’s subgrid dynamical friction model produces more realistic BH orbital evolution and weaker kinetic feedback. We show that increasing feedback strength can recover stronger suppression, but at the cost of tensions with observed AGN luminosity functions and galaxy stellar mass functions. These results pose new challenges for reconciling weak lensing measurements with CMB-inferred cosmology using baryonic feedback alone, potentially implying either novel mechanisms that can efficiently expel gas without compromising other galaxy properties or new non-baryonic physics that reduces matter clustering in the late Universe. Our findings therefore suggest that new theoretical models may be needed to fully exploit Roman weak lensing data.
Co-Authors: Simeon Bird, Yihao Zhou, Tiziana Di Matteo, Rupert Croft, Yueying Ni, and Nianyi Chen
Forecasting the yield of strongly-lensed AGNs in Roman: from quasars to litle red dots
By: Minghao Yue
Type: Talk
Abstract: Strong gravitational lensing enabling critical studies of extragalactic objects that are otherwise inaccessible, thanks to the enhanced spatial resolution and sensitivity offered by lensing magnification. Nevertheless, strong lensing systems are rare and are usually unresolved in ground-based surveys (with Einstein radii <~1"), and surveys for strong lensing systems have been challenging. With unparalleled combination of survey area, PSF sharpness, and depth, Roman will revolutionize the survey and follow-up studies of string lensing systems.
In this talk, I will present the expected yield of lensed AGN surveys in Roman High Latitude Wide Area Survey (HLWAS). I will discuss two especially interesting populations, namely lensed quasars and lensed Little Red Dots (LRDs), including their expected numbers, the survey strategies for these objects, and the follow-up studies they enable. In particular, I will present tools to produce mock catalogs and observations, and demonstrate how the unparalleled power of Roman changes the traditional survey strategy for lensed quasars. I will also highlight studies that can only be achieved by teh Roman-built lensed AGN sample, including (1) probing the small-scale structure of high-redshift IGM and CGM; (2) constraining the SMBH masses and accretion structures of LRDs.
Co-Authors: Xiaohui Fan, Jiani Ding
Toward Robust η⊕ Estimates: : Homogeneous Reanalysis of the Planetary Parameters and Statistical Validation of 188 Kepler Habitable-Zone and Near Habitable Confirmed and Candidate Planets
By: Md Abdullah Al Zaman
Type: Poster
Abstract: Eta-Earth (η⊕) is a crucial concept in the study of the prevalence of Earth-like planets, as well as planetary formation, and the diversity of planetary systems. Precise estimates of η⊕ guide future space8 missions, such as the Habitable World Observatory (HWO) for biosignature detection and the broader search for life beyond Earth. Better stellar and planetary characterization leads to more accurate classification of planets as ‘Earth-like’ and truly within the Habitble Zone (HZ). In this study, we reanalyzed data from 188 confirmed and candidate HZ planets discovered by Kepler around main sequence FGKM stars, with mean radius and period of 2.15 R⊕ and 185 days, respectively, and stellar effective temperatures (Teff ) between 3200 K and 6900 K. Using updated stellar parameters and our Python-based tool, STERNA (Synthetic Transit Lightcurve HarvestER for Exoplanet Science Cases), followed by statistical validation with TRICERATOPS, we derived a homogeneous set of planetary radii, semi-major axes, and incident fluxes. 62 planets in this sample yield false positive probability (FPP) values ≤ 0.10 and ∼ 70% of the planets in the sample yield FPP values within the planet likely range. Our analysis is a step toward reducing misclassification and removing inherent biases in η⊕ estimates. It will also prioritize targets for future space missions, such as JWST, the upcoming Lazuli Space Observatory or other space-based observatories.
Co-Authors: Ian J. M. Crossfield
Exploring the Ha Luminosity Function at z ~ 1.3 and z ~ 2.0 in the CANUCS Lensing Fields with JWST NIRISS
By: Qianqiao Zhou
Type: Poster
Abstract: We present Hα luminosity function (LF) measurements at redshifts z ∼ 1.3 and z ∼ 2.0. We adopted JWST grism data in 5 lensing fields from CANUCS program: ABELL-370, MACS-J0416, MACS-J0417, MACS-J1149, MACS-J1423. Based on emission lines spectroscopically identified in the F115W, F150W and F200W filters, we select 803 Hα emitters from all the redshift bins and fields. Through detailed effective volume and completeness analysis for each source, we construct the Hα LF in two redshift bins. We robustly constrain the faint-end slope of the Hα luminosity function after considering the cosmic variance. The emission-line samples presented here will enable further detailed studies of galaxy properties including metallicities. We find a negligible contribution from bright active galactic nuclei in our sample. The methodology presented here can be readily applicable
to other JWST slitless spectroscopic datasets and future wide-field slitless surveys, including those from Euclid, Roman, and the Chinese Space Station Telescope.
Co-Authors: Xin Wang, Yuxuan Pang, Shengzhe Wang