🔭 astrophysics

Fast X-ray transients in NuSTAR data

By searching the NuSTAR archive for hard X-ray transients, this study identifies five candidate fast X-ray transients, four of which are spectrally distinct from soft-band discoveries and likely represent a common population of low-luminosity gamma-ray bursts with luminosities ranging from $10^{43}$ to $10^{48}$ erg s $^{-1}$ .

Original authors: Murray Brightman, Joahan Castañeda Jaimes, Daniel Stern, Brian Grefenstette

Published 2026-01-22

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Murray Brightman, Joahan Castañeda Jaimes, Daniel Stern, Brian Grefenstette

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe is a giant, dark ocean. For decades, astronomers have been using "soft" flashlights (telescopes that see low-energy X-rays) to look for sudden, bright flashes in this ocean. They call these flashes "Fast X-ray Transients" (FXTs). Sometimes they know what caused them—like a star exploding or two neutron stars crashing—but often, they are mysterious blips with no clear origin.

Recently, a new mission called the Einstein Probe started finding many of these flashes, linking some to massive star deaths and gamma-ray bursts. But there's a catch: these new telescopes only look at the "soft" end of the X-ray spectrum.

This paper is about a team of astronomers who decided to look at the "hard" end of the spectrum using a different telescope called NuSTAR. Think of NuSTAR as a high-powered, hard-X-ray flashlight that sees energy levels the others miss. They wanted to see if there were any mysterious flashes hiding in this harder, brighter light that the soft flashlights couldn't see.

The Big Hunt

The team went through a massive archive of data collected by NuSTAR over 13 years (from 2012 to 2025). It's like sifting through 204 million seconds of video footage, looking for a specific type of event: a flash that lasts about 1,000 seconds (roughly 15 minutes).

Because these flashes are so short, they would normally get lost in the background noise of a long observation. So, the team chopped the data into tiny 1,000-second slices and looked for anything that popped up in just one slice and then vanished.

The Five Suspects

From this massive search, they found five candidate flashes. Here is what makes them special:

They are "Hard" and Bright: Four of the five flashes were "spectrally hard." In simple terms, this means they were made of very high-energy X-rays. Most other known flashes are "soft" (lower energy). These new ones are like high-voltage sparks compared to the gentle glow of the others.
They are Fast: They lasted between 200 and 1,000 seconds.
They are Distant: Three of them seem to come from faraway galaxies, billions of light-years away. This means they are incredibly powerful, releasing as much energy as a massive star explosion.

What Could They Be?

The team tried to figure out what caused these flashes by looking at the surroundings and comparing them to known cosmic events. They ruled out a few ideas:

Not Star Flares: They are too powerful and too "hard" to be simple flares from stars in our own galaxy.
Not Magnetar Explosions: While the energy fits, the duration is too long for the typical "giant flares" from magnetars (super-magnetic neutron stars).
Not Supernova Shock Breakouts: These are the moment a star's shockwave hits the surface. While the timing fits, the "hard" energy of these flashes doesn't quite match the "soft" energy usually seen in these events.

The Leading Theory: Low-Luminosity Gamma-Ray Bursts
The team believes these flashes most likely resemble Low-Luminosity Gamma-Ray Bursts (LLGRBs).

The Analogy: Imagine a standard Gamma-Ray Burst (GRB) as a massive, blinding searchlight that everyone sees. These new flashes are like a dimmer, more common version of that same light. They are less bright, so they are harder to spot, but the team suggests they might actually be much more common in the universe than the bright ones.
The Connection: Since GRBs are often caused by the collapse of massive stars or the collision of compact objects (like neutron stars), these flashes could be the "faint cousins" of those violent cosmic collisions.

Why This Matters

Before this, we mostly knew about these flashes in "soft" X-rays. This paper shows that if you look at the "hard" X-rays, you find a different, rarer, and very energetic population of events.

The authors conclude that these five flashes are likely the first examples of this specific type of hard X-ray transient found in the NuSTAR data. They suggest that the universe is full of these "dimmer" but frequent explosions, and we just needed the right kind of flashlight (NuSTAR) to see them.

In a nutshell: The team used a hard-X-ray telescope to find five mysterious, high-energy flashes in the sky. They aren't the usual suspects; they look like the "underground" version of massive star explosions, suggesting that the universe is full of these powerful, short-lived events that we've been missing until now.

Technical Summary: Fast X-ray Transients in NuSTAR Data

Problem and Motivation
Fast X-ray transients (FXTs) are X-ray flares lasting from hundreds of seconds to a few hours. While known classes include stellar flares, X-ray binary outbursts, supernova shock breakouts, and gamma-ray bursts (GRBs), a population of FXTs with unknown origins has been identified serendipitously in soft X-ray ( $E < 10$ keV) archival data from Chandra and XMM-Newton. Recent discoveries by the Einstein Probe (EP) mission have linked many soft-band FXTs to GRBs and massive star collapses. However, these searches have been limited to soft X-rays. NuSTAR, operating in the 3–79 keV band, is the most sensitive hard X-ray telescope in orbit, yet no systematic search for FXTs had been conducted in its archive. This paper addresses the gap in understanding the hard X-ray population of FXTs and their potential association with high-energy transients.

Methodology
The authors conducted a systematic search of the NuSTAR archive covering observations from its launch in June 2012 through June 2025, totaling 5235 observations and 204 Ms of exposure time.

Data Processing: Observations were processed using HEASOFT and CALDB. To detect short-duration transients that would be diluted in standard long exposures, the data were sliced into overlapping 1000-second time bins (with 500-second steps).
Source Detection: Images were generated in four energy bands (3–79, 3–8, 8–24, and 24–79 keV) by co-adding the FPMA and FPMB detector images. Point sources were identified using the detect function in ximage with a sliding-cell method.
Statistical Thresholds: To account for $\sim 10^8$ independent trials (time and spatial), a false alarm probability threshold of $< 10^{-8}$ was applied.
Candidate Selection: Candidates were required to be detected in only one time slice and located at least 3' from the primary target or other known detections to exclude persistent sources.
Verification: Candidates were cross-matched with existing catalogs (2SXPS, 4XMM, CSC2, 1eRASS) to remove known sources. Remaining candidates underwent manual inspection to rule out stray light and background flares. Spectral and lightcurve analyses were performed using nuproducts and XSPEC (fitting power-law models with Cash/W-statistics).

Key Results
From the 204 Ms of exposure, the authors identified five candidate FXTs:

NuSTAR 150831 (3–79 keV)
NuSTAR 180127 (3–79 keV)
NuSTAR 210321 (24–79 keV)
NuSTAR 210606 (24–79 keV)
NuSTAR 230103 (3–8 keV)

Temporal Properties: The transients have durations of $\sim 200–1000$ s. Lightcurves show peak count rates of 0.15–0.45 counts s $^{-1}$ , with some showing rise-and-fall structures or potential double peaks.
Spectral Properties: Four of the five candidates exhibit spectrally hard spectra with power-law indices in the range $-3 < \Gamma < 0$ . This distinguishes them from the typically softer FXTs discovered in soft X-ray bands. NuSTAR 230103 is the exception, showing a softer spectrum ( $\Gamma \approx 3.2$ ).
Associations and Luminosities:
- Three candidates (NuSTAR 150831, 180127, 230103) have potential associations with galaxies at redshifts $z = 0.1 - 2$ .
- Assuming these associations, the implied 3–79 keV luminosities range from $10^{43}$ to $10^{48}$ erg s $^{-1}$ .
- Two candidates (NuSTAR 210321, 210606) are coincident with the Galactic plane and nearby stars. However, the authors argue against a stellar flare origin due to the high implied luminosities ( $10^{34}–10^{37}$ erg s $^{-1}$ if Galactic), hard spectra, and lack of recurrence.
Event Rates: Based on the potential host galaxies, the authors estimate a volumetric event rate of 125–2900 Gpc $^{-3}$ yr $^{-1}$ .

Significance and Conclusions
The paper concludes that the properties of these NuSTAR FXTs—specifically their durations, hard spectra, and derived volumetric rates—are most consistent with low-luminosity gamma-ray bursts (LLGRBs).

The derived rates are significantly higher than those of high-luminosity GRBs but consistent with the expected population of LLGRBs, suggesting they may be much more common than their high-luminosity counterparts in the redshift range $z \sim 0.1 - 1$ .
One candidate (NuSTAR 230103) is also consistent with a supernova shock breakout (SN SBO), though its hard spectrum contrasts with the soft spectra typically seen in SN SBOs like SN 2008D.
The authors note that these hard X-ray transients would likely be too faint to be detected by the Einstein Probe in the soft band, highlighting the unique capability of NuSTAR to reveal a distinct population of high-energy transients.
The study suggests that while some FXTs may originate from compact object mergers (given the rate consistency with binary neutron star mergers), the hard spectral nature points strongly toward a connection with the GRB phenomenon, specifically the low-luminosity end of the distribution.

The Big Hunt

The Five Suspects

What Could They Be?

Why This Matters

More like this