X-ray Flaring and Variability in NGC 1275, the Heart of the Perseus Cluster

Abstract: NGC 1275 is the central galaxy in the Perseus Cluster. The active galactic nucleus (AGN) within NGC 1275 is notable for its strong and variable radio activity, tied to the production of radio jets that inflate large bubbles in the hot intracluster medium (ICM). High spatial resolution X-ray imaging can separate the AGN from the bright ICM, but monitoring the mass accretion rate onto the black hole and establishing disk-jet connections in NGC 1275 requires a high cadence. Here, we report on X-ray monitoring of NGC 1275 using data taken over 20 years with the Neil Gehrels Swift Observatory. Modeling the temporally constant ICM in each observation allows X-ray emission from accretion onto the black hole to be traced reliably, with typical flux errors of $\sim 3\%$. X-ray flaring by a factor of $\sim2$ over mere days is detected starting on MJD 59956 (2023 Feb. 21). The flares imply an emission region consistent with $$r \leq 870~(10^{{8}~M_{\odot}/M_{BH})~} GM/c<sup>{2}$$. The profile of the flaring is inconsistent with simple predictions for tidal disruption events. A flare appears roughly 300 days later in radio monitoring data at 43 GHz. Overall, our results indicate that coordinated, moderate-resolution X-ray imaging and radio monitoring could potentially trace disk-jet connections in the AGN that most vividly impact large-scale structure, and be extended to other sources that impact their hosts.

• The paper reconstructs a nearly two-decade X-ray light curve from 89 Swift/XRT observations by modeling the constant intracluster medium and measuring AGN fluxes with typical errors of about 3%.
• The paper identifies a 2023 flare that doubled the X-ray flux, produced two peaks lasting roughly five days, and constrained the emitting region to no more than 7.8 × 10^15 cm, supporting a compact corona or jet origin.
• The paper disfavors a tidal disruption event because the flare decays as t^-0.27 and t^-0.11 rather than the canonical t^-5/3 law, while sparse radio monitoring leaves a possible approximately 300-day disk–jet connection unresolved.

Overview

NGC 1275, the dominant central galaxy of the Perseus Cluster and host of the radio source 3C 84, presents a persistent observational challenge: its active galactic nucleus (AGN) is embedded in one of the brightest diffuse X-ray sources in the sky, making it difficult to isolate accretion-driven variability from the static intracluster medium (ICM). Ketchum et al. address this by analyzing 89 Swift/XRT "photon counting" observations spanning roughly 2007–2026, modeling the temporally constant ICM component in each spectrum so that the variable power-law emission from accretion onto the black hole can be tracked with typical flux errors of approximately 3%. The analysis yields a nearly two-decade light curve of the AGN and reveals the strongest X-ray flaring yet observed in this source, beginning on MJD 59956 (2023 February 21), with flux rising by a factor of approximately 2 over timescales of days.

Methodology

The authors extracted spectra using an 18 arcsecond circular region centered on the X-ray nucleus, deliberately omitting background subtraction because the diffuse cluster emission varies continuously with radius; the goal is relative flux characterization rather than absolute ICM measurement. All spectra were optimally binned following Kaastra & Bleeker and fit in XSPEC with Cash statistics over the 0.3–10.0 keV band.

A composite model, tbabs*zmshift*(apec+pow), was first applied to the summed spectrum of the first 40 observations (chosen to exclude known flares). This yielded a power-law index $\Gamma = 1.68^{+0.06}_{-0.04}$ — fully consistent with the canonical Seyfert value of 1.7, which was then frozen — a column density $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$, and an ICM temperature $kT = 5.2 \pm 0.1$ keV. The authors acknowledge that this single-temperature plasma treatment is a simplification: Hitomi and XRISM spectroscopy demonstrate that the Perseus core requires temperature gradients, velocity structure, and abundance variations. Nevertheless, the simple model adequately captures the strongest lines, including the 6.7 keV Fe XXV complex that could otherwise bias the power-law normalization.

In fits to individual observations, the ICM parameters were frozen and only the power-law normalization varied. The resulting mean unabsorbed flux is $5.40 \times 10^{-11}\ {\rm erg\ cm^{-2}\ s^{-1}}$, corresponding to a mean luminosity of $3.76 \times 10^{43}\ {\rm erg\ s^{-1}}$ at the luminosity distance of 76.3 Mpc.

The 2023 flare

The dominant feature of the light curve is a flare peaking near MJD 60000, lasting less than 60 days (MJD 59930–59990) and composed of at least two distinct peaks with characteristic durations of approximately 5 days each. The brightest single observation reaches $11.34 \times 10^{-11}\ {\rm erg\ cm^{-2}\ s^{-1}}$, roughly twice the long-term median. Verification was performed in two ways: direct comparison of the brightest observation to the summed spectrum shows the difference is consistent with a pure absorbed power-law, and a difference spectrum fit in SPEX gives an acceptable statistic ($C = 97.3$ for $\nu = 94$).

The approximately 5-day variability timescale implies an emission region of $r \leq c\Delta t \approx 7.8 \times 10^{15}$ cm, or $r \leq 870\,(10^8 M_\odot/M_{\rm BH})\ GM/c^2$. Combined with the Seyfert-like spectral index of the variable component, this is consistent with a compact coronal region of the kind invoked in magnetically arrested disk (MAD) jet-launching scenarios. However, the authors are explicit that the data do not definitively rule out a downstream origin within the relativistic jet.

Excluding a tidal disruption origin

Because some black hole mass estimates for NGC 1275 fall below the $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$0 threshold above which main-sequence stars are swallowed whole, a tidal disruption event (TDE) is nominally testable against the canonical $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$1 decay. The observed decays are markedly slower: the stronger peak is best fit by $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$2 and the weaker peak by $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$3. The TDE interpretation is therefore disfavored, leaving fluctuations in the mass accretion rate or downstream shocks in the jet as the more plausible drivers.

Radio comparison and disk–jet coupling

Cross-correlation with MOJAVE (15 GHz) and VLBA-BU-BLAZAR (43 GHz) monitoring reveals comparable fractional variability across bands: 28.57% in X-rays, 28.09% at 15 GHz, and 28.83% at 43 GHz over the full baseline. Notably, when the flare interval is excluded (MJD 55000–59000), the fractional variability drops to 18.17% (X-ray), 19.40% (15 GHz), but remains high at 27.60% (43 GHz), suggesting the 43 GHz emission varies independently of the X-ray state.

Two features bear on disk–jet coupling. First, the radio points immediately preceding the X-ray flare lie below their local mean and median values — a pattern reminiscent of the inverse X-ray/radio correlations seen in GRS 1915+105, Cygnus X-1, NGC 4051, and 3C 120, where radio flares follow dips interpreted as ejection of the inner flow. Second, 43 GHz flux density rises above 4 Jy starting at MJD 60230, approximately 296 days after the onset of X-ray flaring. This lag could represent propagation from the compact corona to a radio-emitting zone in the jet, or between two downstream regions. The authors appropriately caution that sparse sampling in all bands limits the strength of these associations; no firm claim of a causal connection is made.

Constraints on the black hole mass

The mass of the black hole remains contested: stellar dynamical estimates give $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$4 (Scharwächter et al.) and $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$5 (Riffel et al.), while a Pa$$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$6 scaling relation yields $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$7, far below the $$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$8–$$N_{\rm H} = 1.99 \pm 0.04 \times 10^{21}\ {\rm cm^{-2}}$$9 relation. The bolometric luminosity inferred here constrains this debate: if $kT = 5.2 \pm 0.1$0, the implied Eddington fraction is roughly 0.3 after applying standard bolometric corrections, which is high for a jet-producing system. A lower Eddington fraction would favor a larger mass, potentially $kT = 5.2 \pm 0.1$1. Tentative evidence that inverse disk–jet coupling emerges above an Eddington fraction of 0.1 similarly suggests the mass does not greatly exceed $kT = 5.2 \pm 0.1$2.

Three circumstantial arguments favor an accretion-flow origin for the flares: the Seyfert-like $kT = 5.2 \pm 0.1$3 index of the variable component; the requirement of isotropic X-ray illumination to produce the narrow Fe K$kT = 5.2 \pm 0.1$4 line confirmed by Hitomi and confined to <100 pc by Chandra imaging; and the compactness limit consistent with a corona capable of illuminating the accretion flow. The authors concede, however, that flares accompanied by contemporaneous radio and $kT = 5.2 \pm 0.1$5-ray increases may still arise downstream in the jet.

Limitations and open questions

Several limitations qualify the results. The light curve is sparsely and erratically sampled, including a gap of nearly 1500 days between MJD 58455 and 59931, and only one observation in the MJD 58000–59900 window; additional flaring during gaps cannot be excluded. The simplified single-temperature ICM model, while adequate for isolating the power-law component, does not capture the full complexity demonstrated by microcalorimeter spectroscopy. The approximately 300-day X-ray-to-radio lag rests on coarse sampling and remains unconfirmed. Finally, the degeneracy between a compact-corona origin and downstream jet emission for individual flares is unresolved by the present data.

Conclusion

This work demonstrates that moderate-resolution X-ray imaging with Swift, combined with fixed modeling of the ICM contribution, can recover AGN variability in NGC 1275 at the few-percent flux level despite severe contamination from the Perseus Cluster. The detection of factor-of-two flaring on day-scale timescales — inconsistent with TDE decay laws and consistent with a compact accretion-flow origin — together with a possible ~300-day radio counterpart, establishes coordinated X-ray and radio monitoring as a viable path toward tracing disk–jet connections in cluster-center AGN. The authors further note a practical benefit for cluster science: identifying periods of low AGN activity through such monitoring would improve XRISM plasma diagnostics in cluster cores, where AGN continuum contamination currently degrades abundance measurements even though turbulent pressure measurements (4–6% of thermal in the innermost 20 kpc) are minimally affected.