NICER observations reveal doubled timescales in Ansky's quasi-periodic eruptions (QPEs)

TL;DR

This study analyzes 2025 NICER observations of Ansky's quasi-periodic eruptions (QPEs) and compares them to 2024 to test emission models for QPEs near supermassive black holes. The data reveal QPEs four times more energetic, with $T_{rec} \sim 10$ days and $T_{dur} \sim 2.5$–$4$ days, and a systematic increase in recurrence time within the 2025 season, quantified as $\dot P \approx +0.02$ between years and $\approx +0.1$ days per flare within 2025; flares also become more asymmetric with longer decays. Time-resolved spectroscopy shows persistent $L_{bol}-T$ hysteresis and expanding $R_{bb}$ during eruptions, consistent with an evolving emitting region. The authors propose a qualitative EMRI/debris-disk interaction scenario where 2024 featured two QPEs per orbit due to debris penetrating the disk, while 2025 favors one-sided shocks with reduced penetration, though mass-transfer and magnetized disk-instability models remain viable alternatives requiring further numerical tests. Continued NICER and XMM-Newton monitoring will be essential to constrain the physical mechanism and its time evolution.

Abstract

Quasi-periodic eruptions (QPEs) are recurring X-ray bursts originating from the vicinity of supermassive black holes, but their driving mechanisms remain under debate. This study analyzes new NICER observations of QPEs in Ansky (a transient event in the nucleus of the galaxy SDSS J1335+0728), taken between January and June 2025. By examining flare durations, peak-to-peak recurrence times, and profiles, we compare the 2025 data with those from 2024 to investigate changes in energy, timescales, and flare shapes. The 2025 QPEs are found to be four times more energetic, with recurrence times of approximately 10 days and flare durations ranging from 2.5 to 4 days, making them both about twice as long as in 2024. Additionally, the flare profiles have become more asymmetric, showing longer decays. We explore different theoretical scenarios to explain the observed properties of the QPEs in Ansky, including evolving stream-disk interactions in an extreme mass-ratio inspiral (EMRI) system as a potential mechanism behind the observed changes in recurrence time and energetics, while also considering alternative models based on mass transfer and accretion disk instabilities. Continued observational efforts will be crucial for unveiling the nature of Ansky.

Figures (6)

• Figure 1: NICER X-ray light curve of Ansky for the period (upper panel) May 19--July 20, 2024. Detections are plotted as black circles, and non-detections as gray triangles. The typical peak-to-peak timescale is $\sim$4.5 days; these QPEs were presented in Hernandez25. NICER light curve (bottom panel) between January 7 and March 30, 2025. The typical peak-to-peak timescale is $\sim$10 days. Note that there are no observations around MJD 60700, so there might be a missing flare. The dashed lines represent exponential rise and decay profiles fit to each QPE (see Sect. \ref{['results']}). The horizontal axis spans an equal number of days in 2024 and 2025 to facilitate timescale comparison.
• Figure 2: Comparison of the flare profiles from 2024 (left panel) to 2025 (right panel).
• Figure 3: Time between consecutive flares as a function of flare number. Error bars reflect uncertainties propagated from the flare peak times. A linear fit (red dashed line) shows a gradual increase in recurrence time, rising by about 0.1 days per flare.
• Figure 4: Background-subtracted X-ray spectrum from NICER. This is an example of a high-flux spectrum, taken from one 200-second exposure, used to estimate the spectral parameters. The spectrum is fitted with a blackbody model. The residuals in the bottom panel show the presence of an additional component, corresponding to the emission/absorption feature around $\sim$1 keV, analyzed in detail in Chakraborty25b.
• Figure 5: The spectral evolution of the QPEs in Ansky for the bursts observed by NICER in 2025. The color scheme indicates the time relative to peak (dark/light at early/late times). The QPEs undergo hysteresis in the $L$-$kT$ plane (left panel), which, for a blackbody-like spectrum, can be interpreted as an expanding emission region $R_{\mathrm{bb}}$ (right panel). The error bars represent $\pm 1\sigma$ uncertainties.