NASA’s Nancy Grace Roman Space Telescope is expected to detect roughly 100 tidal disruption events per year, each one the signature of a supermassive black hole ripping apart a star. Many of these events will come from a period known as cosmic noon, about 11 to 12 billion years ago, when galaxies were forming stars and feeding their central black holes at the highest rates in cosmic history. The telescope’s High-Latitude Time-Domain Survey, or HLTDS, will repeatedly image wide patches of sky to catch the slow brightening and fading of these distant stellar deaths, reaching redshifts that no existing survey can reliably probe.
Why catching ancient stellar destruction matters right now
Current ground-based surveys can spot tidal disruption events, or TDEs, only in the relatively nearby universe. The Zwicky Transient Facility, for example, has built a spectroscopically complete sample that defines the optical luminosity function and volumetric rates astronomers use as a baseline. Those rates describe how often black holes shred stars in today’s galaxies, but they say little about conditions billions of years ago, when galaxy mergers were far more frequent and nuclear stellar densities were likely much higher.
Roman’s HLTDS is designed to close that gap. By scanning the same fields season after season in near-infrared light, the survey will capture TDE light curves at redshifts well beyond 1, corresponding to lookback times stretching past 8 billion years and, for the brightest events, into the cosmic noon window around redshift 3 and beyond. That reach matters because the rate at which black holes destroy stars depends on how many stars crowd into a galaxy’s core, and mergers during cosmic noon should have packed those cores far more tightly than anything seen in local galaxies.
A testable prediction follows from that logic. If Roman’s first-year TDE harvest includes a disproportionate share of events in post-merger host galaxies at high redshift, compared to the merger fractions seen in ZTF’s local sample, it would be direct evidence that elevated nuclear stellar densities during cosmic noon drove higher disruption rates. That signal would connect black hole feeding to the broader story of galaxy assembly in a way that no current dataset can.
Forecast models and the peer-reviewed evidence behind Roman’s TDE yield
The quantitative backbone for these expectations comes from a peer-reviewed study by Karmen et al., published in The Astrophysical Journal, which forecasts TDE detection rates as a function of redshift for Roman’s HLTDS, the Vera C. Rubin Observatory’s Legacy Survey of Space and Time (LSST), and serendipitous detections by JWST’s COSMOS-Web program. The paper models how galaxy nuclear stellar densities, merger rates, and dust obscuration evolve with redshift, then folds those ingredients through each telescope’s survey parameters to predict annual yields.
For Roman specifically, the forecast lands at approximately 100 TDEs per year, a figure that NASA highlights in its own mission planning materials. Rubin/LSST will find large numbers of TDEs as well, but primarily at lower redshifts because it operates in optical wavelengths that fade more quickly with distance. Roman’s near-infrared sensitivity gives it a distinct advantage for events beyond redshift 2, where optical light from a disrupted star has been stretched into infrared bands by the expansion of the universe.
An independent cross-check comes from the Hourglass simulation catalog, which models transient yields for Roman’s time-domain survey under explicit signal-to-noise thresholds and a science-agnostic detection pipeline. While the Hourglass work uses different assumptions about survey cadence and detection criteria, it broadly supports the expectation that Roman will recover a statistically meaningful sample of high-redshift TDEs, not just a handful of marginal detections. Together, these modeling efforts suggest that Roman will move the field from isolated case studies toward population-level statistics at early cosmic times.
Open questions Roman’s TDE census still needs to answer
No telescope has yet detected a confirmed TDE at the redshifts Roman is targeting. Every high-redshift yield number in the current literature rests on extrapolated models, not empirical detections. The Karmen et al. forecasts depend on assumptions about how nuclear stellar densities and black hole occupation fractions evolve with redshift, and those assumptions carry significant uncertainty. If the true occupation fraction of intermediate-mass black holes in early dwarf galaxies is lower than modeled, or if dust obscuration at cosmic noon is heavier than expected, the actual yield could fall well short of 100 events per year.
Specific operational details also remain partially defined. The exact per-field cadence, filter selection, and signal-to-noise thresholds that will govern TDE identification in the HLTDS pipeline have been described at a high level in NASA planning documents, but the final survey design is still being refined. Cadence choices matter enormously: TDE light curves rise and fall over weeks to months, and missing the peak or catching only the tail of an event can make classification ambiguous, especially when the host galaxy is faint and distant. Optimizing the trade-off between sky coverage and visit frequency will be critical to meeting the forecast yields.
Another open issue is how cleanly Roman will be able to distinguish TDEs from other nuclear transients. Supernovae that explode very close to galaxy centers, active galactic nucleus (AGN) flares, and microlensing events can all mimic some aspects of a TDE light curve. Karmen et al. assume that multi-band photometry and characteristic rise and decay times will allow robust classification, but those assumptions have been calibrated mainly on low-redshift events. At higher redshift, lower signal-to-noise data and stronger host contamination may blur those distinctions.
Addressing these challenges will require coordinated follow-up from other observatories. X-ray and radio observations, for instance, can reveal high-energy emission and jets that are more typical of TDEs than of ordinary supernovae. Spectroscopy from large ground-based telescopes can search for the broad emission lines and temperature evolution associated with stellar debris falling onto a black hole. Forecasts for this multi-wavelength ecosystem are being explored in work such as the Roman transient simulations, which examine how different follow-up strategies affect classification confidence and science return.
From stellar deaths to black hole growth
If Roman does approach its expected TDE yield, the resulting dataset will address several long-standing questions about black hole growth. One is whether tidal disruption contributes significantly to building up the masses of supermassive black holes in low-mass galaxies. In the nearby universe, TDEs are rare enough that they seem like a minor growth channel, but if disruption rates were much higher during cosmic noon, they could have played a more substantial role in feeding nascent black holes.
Another question concerns the demographics of black holes that can produce observable TDEs. Very massive black holes, above a few hundred million solar masses, are expected to swallow stars whole without generating a bright flare. By mapping how the TDE rate changes with host galaxy mass and redshift, Roman can test whether this theoretical cutoff holds at early times and whether intermediate-mass black holes were more common in young galaxies than they are today.
Finally, Roman’s TDE census will provide a new probe of the connection between galaxy mergers and nuclear activity. If the survey finds that disruption rates peak in morphologically disturbed galaxies, or in systems with close companions, it would bolster the idea that mergers are efficient at driving stars and gas into galactic centers. Conversely, a weak or absent correlation would push theorists to rethink how black holes and their host galaxies co-evolve, especially in the tumultuous era around cosmic noon.
In all of these areas, Roman will not work in isolation. Its high-redshift TDE discoveries will be combined with the rich low-redshift samples from ZTF and, soon, LSST to build a continuous picture of how stellar destruction by black holes has changed over the age of the universe. By extending TDE studies into territory that has so far been accessible only in simulations and models, the Nancy Grace Roman Space Telescope promises to turn the violent deaths of stars into precise tools for tracing the growth of the darkest objects in the cosmos.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
