Relativistic reverberation in the accretion flow of a tidal disruption event


Our current understanding of the curved space-time around supermassive black holes is based on actively accreting black holes, which make up only ten per cent or less of the overall population. X-ray observations of that small fraction reveal strong gravitational redshifts that indicate that many of these black holes are rapidly rotating1; however, selection biases suggest that these results are not necessarily reflective of the majority of black holes in the Universe2. Tidal disruption events, where a star orbiting an otherwise dormant black hole gets tidally shredded and accreted onto the black hole3, can provide a short, unbiased glimpse at the space-time around the other ninety per cent of black holes. Observations of tidal disruptions have hitherto revealed the formation of an accretion disk and the onset of an accretion-powered jet4,5,6,7,8, but have failed to reveal emission from the inner accretion flow, which enables the measurement of black hole spin. Here we report observations of reverberation9,10,11,12 arising from gravitationally redshifted iron Kα photons reflected off the inner accretion flow in the tidal disruption event Swift J1644+57. From the reverberation timescale, we estimate the mass of the black hole to be a few million solar masses, suggesting an accretion rate of 100 times the Eddington limit or more13. The detection of reverberation from the relativistic depths of this rare super-Eddington event demonstrates that the X-rays do not arise from the relativistically moving regions of a jet, as previously thought5,14.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Lag–energy spectrum of Swift J1644+57.
Figure 2: The energy spectrum of Swift J1644+57.
Figure 3: Comparison of lags in Swift J1644+57 with other Seyfert galaxies.
Figure 4: Schematic of reverberation off a super-Eddington accretion flow.


  1. 1

    Reynolds, C. S. The spin of supermassive black holes. Class. Quant. Grav. 30, 244004 (2013)

    Article  ADS  MathSciNet  MATH  Google Scholar 

  2. 2

    Vasudevan, R. V., Fabian, A. C., Reynolds, C. S., Dauser, T. & Gallo, L. C. A selection effect boosting the contribution from rapidly spinning black holes to the cosmic X-ray background. Mon. Not. R. Astron. Soc. 458, 2012–2023 (2016)

    CAS  Article  ADS  Google Scholar 

  3. 3

    Rees, M. J. Tidal disruption of stars by black holes of 106–108 solar masses in nearby galaxies. Nature 333, 523–528 (1988)

    Article  ADS  Google Scholar 

  4. 4

    Burrows, D. N. et al. Relativistic jet activity from the tidal disruption of a star by a massive black hole. Nature 476, 421–424 (2011)

    CAS  Article  ADS  Google Scholar 

  5. 5

    Zauderer, B. A. et al. Birth of a relativistic outflow in the unusual gamma-ray transient Swift J164449.3+573451. Nature 476, 425–428 (2011)

    CAS  Article  ADS  Google Scholar 

  6. 6

    Bloom, J. S. et al. A possible relativistic jetted outburst from a massive black hole fed by a tidally disrupted star. Science 333, 203–206 (2011)

    CAS  Article  ADS  Google Scholar 

  7. 7

    Miller, J. M. et al. Flows of X-ray gas reveal the disruption of a star by a massive black hole. Nature 526, 542–545 (2015)

    CAS  Article  ADS  Google Scholar 

  8. 8

    Komossa, S. Tidal disruption of stars by supermassive black holes: status of observations. J. High Energy Astrophys. 7, 148–157 (2015)

    Article  ADS  Google Scholar 

  9. 9

    Fabian, A. C. et al. Broad line emission from iron K- and L-shell transitions in the active galaxy 1H 0707-495. Nature 459, 540–542 (2009)

    CAS  Article  ADS  Google Scholar 

  10. 10

    Zoghbi, A., Fabian, A. C., Reynolds, C. S. & Cackett, E. M. Relativistic iron K X-ray reverberation in NGC 4151. Mon. Not. R. Astron. Soc. 422, 129–134 (2012)

    Article  ADS  Google Scholar 

  11. 11

    Cackett, E. M. et al. Modelling the broad Fe Kα reverberation in the AGN NGC 4151. Mon. Not. R. Astron. Soc. 438, 2980–2994 (2014)

    CAS  Article  ADS  Google Scholar 

  12. 12

    Uttley, P., Cackett, E. M., Fabian, A. C., Kara, E. & Wilkins, D. R. X-ray reverberation around accreting black holes. Astron. Astrophys. Rev. 22, 72 (2014)

    Article  ADS  Google Scholar 

  13. 13

    Evans, C. R. & Kochanek, C. S. The tidal disruption of a star by a massive black hole. Astrophys. J. 346, L13–L16 (1989)

    Article  ADS  Google Scholar 

  14. 14

    Tchekhovskoy, A., Metzger, B. D., Giannios, D. & Kelley, L. Z. Swift J1644+57 gone MAD: the case for dynamically important magnetic flux threading the black hole in a jetted tidal disruption event. Mon. Not. R. Astron. Soc. 437, 2744–2760 (2014)

    Article  ADS  Google Scholar 

  15. 15

    Levan, A. J. et al. An extremely luminous panchromatic outburst from the nucleus of a distant galaxy. Science 333, 199–202 (2011)

    CAS  Article  ADS  Google Scholar 

  16. 16

    Coughlin, E. R. & Begelman, M. C. Hyperaccretion during tidal disruption events: weakly bound debris envelopes and jets. Astrophys. J. 781, 82 (2014)

    Article  ADS  Google Scholar 

  17. 17

    Reis, R. C. et al. A 200-second quasi-periodicity after the tidal disruption of a star by a dormant black hole. Science 337, 949–951 (2012)

    CAS  Article  ADS  Google Scholar 

  18. 18

    Zoghbi, A. et al. Discovery of Fe Kα X-ray reverberation around the black holes in MCG-5-23-16 and NGC 7314. Astrophys. J. 767, 121 (2013)

    Article  ADS  CAS  Google Scholar 

  19. 19

    Kara, E. et al. Discovery of high-frequency iron K lags in Ark 564 and Mrk 335. Mon. Not. R. Astron. Soc. 434, 1129–1137 (2013)

    CAS  Article  ADS  Google Scholar 

  20. 20

    Alston, W. N. et al. Discovery of a ~2-h high-frequency X-ray QPO and iron Kα reverberation in the active galaxy MS 2254.9−3712. Mon. Not. R. Astron. Soc. 449, 467–476 (2015)

    CAS  Article  ADS  Google Scholar 

  21. 21

    De Marco, B. et al. Discovery of a relation between black hole mass and soft X-ray time lags in active galactic nuclei. Mon. Not. R. Astron. Soc. 431, 2441–2452 (2013)

    Article  ADS  Google Scholar 

  22. 22

    Ravasio, M. et al. Observing Mkn 421 with XMM-Newton: the EPIC-PN point of view. Astron. Astrophys. 424, 841–855 (2004)

    CAS  Article  ADS  Google Scholar 

  23. 23

    Sądowski, A. & Narayan, R. Powerful radiative jets in supercritical accretion discs around non-spinning black holes. Mon. Not. R. Astron. Soc. 453, 3213–3221 (2015)

    Article  ADS  Google Scholar 

  24. 24

    Jiang, Y.-F., Stone, J. M. & Davis, S. W. A global three-dimensional radiation magneto-hydrodynamic simulation of super-Eddington accretion disks. Astrophys. J. 796, 106 (2014)

    Article  ADS  Google Scholar 

  25. 25

    McKinney, J. C., Dai, L. & Avara, M. J. Efficiency of super-Eddington magnetically-arrested accretion. Mon. Not. R. Astron. Soc. 454, L6–L10 (2015)

    Article  ADS  Google Scholar 

  26. 26

    Abramowicz, M. A., Czerny, B., Lasota, J. P. & Szuszkiewicz, E. Slim accretion disks. Astrophys. J. 332, 646–658 (1988)

    Article  ADS  Google Scholar 

  27. 27

    Abramowicz, M. A. et al. Leaving the innermost stable circular orbit: the inner edge of a black-hole accretion disk at various luminosities. Astron. Astrophys. 521, A15 (2010)

    Article  Google Scholar 

  28. 28

    Krolik, J. H. & Hawley, J. F. Where is the inner edge of an accretion disk around a black hole? Astrophys. J. 573, 754–763 (2002)

    Article  ADS  Google Scholar 

  29. 29

    Jansen, F. et al. XMM-Newton observatory. I: the spacecraft and operations. Astron. Astrophys. 365, L1–L6 (2001)

    Article  ADS  Google Scholar 

  30. 30

    Strüder, L. et al. The European Photon Imaging Camera on XMM-Newton: the pn-CCD camera. Astron. Astrophys. 365, L18–L26 (2001)

    Article  ADS  Google Scholar 

  31. 31

    Zoghbi, A. & Fabian, A. C. X-ray reverberation close to the black hole in RE J1034+396. Mon. Not. R. Astron. Soc. 418, 2642–2647 (2011)

    Article  ADS  Google Scholar 

  32. 32

    Nowak, M. A., Vaughan, B. A., Wilms, J., Dove, J. B. & Begelman, M. C. Rossi x-ray timing explorer observation of Cygnus X-1. II. Timing analysis. Astrophys. J. 510, 874–891 (1999)

    Article  ADS  Google Scholar 

  33. 33

    Zoghbi, A., Reynolds, C. & Cackett, E. M. Calculating time lags from unevenly sampled light curves. Astrophys. J. 777, 24 (2013)

    Article  ADS  Google Scholar 

  34. 34

    Zoghbi, A. et al. Observations of MCG-5-23-16 with Suzaku, XMM-Newton and NuSTAR: disk tomography and Compton hump reverberation. Astrophys. J. 789, 56 (2014)

    Article  ADS  CAS  Google Scholar 

  35. 35

    Kara, E. et al. Iron K and Compton hump reverberation in SWIFT J2127.4+5654 and NGC 1365 revealed by NuSTAR and XMM-Newton. Mon. Not. R. Astron. Soc. 446, 737–749 (2015)

    CAS  Article  ADS  Google Scholar 

  36. 36

    Timmer, J. & König, M. On generating power law noise. Astron. Astrophys. 300, 707–710 (1995)

    ADS  Google Scholar 

  37. 37

    Saxton, C. J., Soria, R., Wu, K. & Kuin, N. P. M. Long-term X-ray variability of Swift J1644+57. Mon. Not. R. Astron. Soc. 422, 1625–1639 (2012)

    Article  ADS  Google Scholar 

  38. 38

    Kass, R. & Raftery, A. Bayes factors. J. Am. Stat. Assoc. 90, 773–795 (1995)

    Article  MathSciNet  MATH  Google Scholar 

  39. 39

    Grier, C. J. et al. Stellar velocity dispersion measurements in high-luminosity quasar hosts and implications for the AGN black hole mass scale. Astrophys. J. 773, 90 (2013)

    Article  ADS  Google Scholar 

  40. 40

    Kaspi, S. Reverberation measurements for 17 quasars and the size-mass-luminosity relations in active galactic nuclei. Astrophys. J. 533, 631–649 (2000)

    CAS  Article  ADS  Google Scholar 

  41. 41

    Bentz, M. C. et al. The radius-luminosity relationship for active galactic nuclei: the effect of host-galaxy starlight on luminosity measurements. II. The full sample of reverberation-mapped AGNs. Astrophys. J. 697, 160–181 (2009)

    CAS  Article  ADS  Google Scholar 

  42. 42

    Bentz, M. C. et al. The low-luminosity end of the radius-luminosity relationship for active galactic nuclei. Astrophys. J. 767, 149 (2013)

    Article  ADS  CAS  Google Scholar 

  43. 43

    Protassov, R., van Dyk, D. A., Connors, A., Kashyap, V. L. & Siemiginowska, A. Statistics, handle with care: detecting multiple model components with the likelihood ratio test. Astrophys. J. 571, 545–559 (2002)

    Article  ADS  Google Scholar 

  44. 44

    Guillochon, J., McCourt, M., Chen, X., Johnson, M. D. & Berger, E. Unbound debris streams and remnants resulting from the tidal disruptions of stars by supermassive black holes. Preprint at (2015)

  45. 45

    Shiokawa, H., Krolik, J. H., Cheng, R. M., Piran, T. & Noble, S. C. General relativistic hydrodynamic simulation of accretion flow from a stellar tidal disruption. Astrophys. J. 804, 85 (2015)

    Article  ADS  Google Scholar 

  46. 46

    Dai, L., McKinney, J. C. & Miller, M. C. Soft X-ray temperature tidal disruption events from stars on deep plunging orbits. Astrophys. J. 812, L39 (2015)

    Article  ADS  CAS  Google Scholar 

  47. 47

    Shakura, N. I. & Sunyaev, R. A. Black holes in binary systems: observational appearance. Astron. Astrophys. 24, 337–355 (1973)

    ADS  Google Scholar 

  48. 48

    Bentz, M. C. et al. A reverberation-based mass for the central black hole in NGC 4151. Astrophys. J. 651, 775–781 (2006)

    CAS  Article  ADS  Google Scholar 

  49. 49

    Kara, E., Fabian, A. C., Cackett, E. M., Miniutti, G. & Uttley, P. Revealing the X-ray source in IRAS 13224-3809 through flux-dependent reverberation lags. Mon. Not. R. Astron. Soc. 430, 1408–1413 (2013)

    CAS  Article  ADS  Google Scholar 

  50. 50

    Bian, W. & Zhao, Y. On X-ray variability in narrow-line and broad-line active galactic nuclei. Mon. Not. R. Astron. Soc. 343, 164–168 (2003)

    CAS  Article  ADS  Google Scholar 

  51. 51

    Kara, E. et al. The closest look at 1H0707-495: X-ray reverberation lags with 1.3 Ms of data. Mon. Not. R. Astron. Soc. 428, 2795–2804 (2013)

    Article  ADS  Google Scholar 

  52. 52

    Zhou, X.-L. & Wang, J.-M. Narrow iron K lines in active galactic nuclei: evolving populations? Astrophys. J. 618, L83–L86 (2005)

    CAS  Article  ADS  Google Scholar 

  53. 53

    Zhang, E.-P. & Wang, J.-M. The unified model of active galactic nuclei. I: non-hidden broad-line region Seyfert 2 and narrow-line Seyfert 1 galaxies. Astrophys. J. 653, 137–151 (2006)

    CAS  Article  ADS  Google Scholar 

  54. 54

    Grier, C. J. et al. A reverberation lag for the high-ionization component of the broad-line region in the narrow-line Seyfert 1 Mrk 335. Astrophys. J. 744, L4 (2012)

    Article  ADS  CAS  Google Scholar 

  55. 55

    Kara, E., Cackett, E. M., Fabian, A. C., Reynolds, C. & Uttley, P. The curious time lags of PG 1244+026: discovery of the iron K reverberation lag. Mon. Not. R. Astron. Soc. 439, L26–L30 (2014)

    CAS  Article  ADS  Google Scholar 

  56. 56

    Jin, C., Done, C., Middleton, M. & Ward, M. A long XMM-Newton observation of an extreme narrow-line Seyfert 1: PG 1244+026. Mon. Not. R. Astron. Soc. 436, 3173–3185 (2013)

    Article  ADS  Google Scholar 

  57. 57

    Marinucci, A. et al. Simultaneous NuSTAR and XMM-Newton 0.5–80 keV spectroscopy of the narrow-line Seyfert 1 galaxy SWIFT J2127.4+5654. Mon. Not. R. Astron. Soc. 440, 2347–2356 (2014)

    CAS  Article  ADS  Google Scholar 

  58. 58

    Malizia, A. et al. First high-energy observations of narrow-line Seyfert 1s with INTEGRAL/IBIS. Mon. Not. R. Astron. Soc. 389, 1360–1366 (2008)

    CAS  Article  ADS  Google Scholar 

  59. 59

    Simien, F. & Prugniel, P. Kinematical data on early-type galaxies. VI. Astron. Astrophys. 384, 371–382 (2002)

    Article  ADS  Google Scholar 

  60. 60

    Alston, W. N., Vaughan, S. & Uttley, P. The flux-dependent X-ray time lags in NGC 4051. Mon. Not. R. Astron. Soc. 435, 1511–1519 (2013)

    Article  ADS  Google Scholar 

  61. 61

    Denney, K. D. et al. A revised broad-line region radius and black hole mass for the narrow-line Seyfert 1 NGC 4051. Astrophys. J. 702, 1353–1366 (2009)

    CAS  Article  ADS  Google Scholar 

Download references


E.K. thanks A. Zoghbi, M. C. Miller, F. Tombesi, E. Miller and L. Denby for discussions. E.K. also thanks the Hubble Fellowship Program for support under grant number HST-HF2-51360.001-A from the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Incorporated, under NASA contract NAS5-26555. J.M.M. acknowledges N. Schartel and XMM-Newton for executing target-of-opportunity observations of Swift J1644+57. C.R. acknowledges support from NASA under grant number NNX14AF86G. L.D. thanks J. McKinney for discussions. L.D. acknowledges support from NASA/NSF/TCAN (NNX14AB46G), NSF/XSEDE/TACC (TG- PHY120005) and NASA/Pleiades (SMD-14-5451). This work is based on observations made with XMM-Newton, a European Space Agency (ESA) science mission with instruments and contributions directly funded by ESA member states and the US (NASA) and the Suzaku satellite, a collaborative mission between the space agencies of Japan (JAXA) and the US (NASA).

Author information




E.K. led the XMM-Newton and Suzaku time lag analysis, simulations, interpretation of the results and manuscript preparation. J.M.M. performed the XMM-Newton spectral analysis and contributed to the interpretation of the results. C.R. developed the analytical toy model for reverberation in a super-Eddington flow and contributed to the interpretation of the results. L.D. had the idea of examining X-ray time lags in a TDE and contributed to the interpretation of the results.

Corresponding author

Correspondence to Erin Kara.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 XMM-Newton lag–frequency spectrum.

The lag–frequency spectrum between rest frame 0.3–1 keV and 1–4 keV (the usual soft and hard bands chosen for reverberation studies9,21). The lag–frequency spectrum shows a negative lag (that is, the soft band lagging behind the hard band) from (2–10) × 10−4 Hz (highlighted by the red hashed region). Because the lag–frequency spectrum shows a soft lag at these frequencies, we examined further the lag–energy spectrum at this particular frequency range (see Fig. 1). Error bars are at the 1σ level.

Extended Data Figure 2 XMM-Newton light curve of Swift J1644+57.

The observed 0.3–10 keV light curve in 200 s bins showing the rapid variability in this 25 ks observation. Note that the y axis does not extend to zero count rate. Error bars are at the 1σ level. Inset, the 4–6 keV (~5.5–8 keV rest frame) light curve (red) and the 6–10 keV (~8–13 keV rest frame) light curve (blue) zoomed in on the most variable part of the light curve from 5,000 s to 15,000 s (indicated by the red line in the main figure). The y axis is in units of count rate divided by the mean of the entire light curve. This illustrates (in the time domain) that on average the 6–10 keV variability leads the 4–6 keV variability. We emphasize that the time delay seen in these coarsely binned light curves places an upper limit on the actual amplitude of the lag. The detailed Fourier analysis (Fig. 1) allows us to obtain a better estimate of the average time lag.

Extended Data Figure 3 Suzaku XIS and HXD spectra.

A ratio plot of the time-integrated energy spectra of the front-illuminated (black) and back-illuminated (red) XIS detectors and the HXD (green) to an absorbed power-law fit from 3–6 keV. The blue line at a ratio of one indicates where the data would be fully described by the absorbed power-law model. Error bars are at the 1σ level. This fit exhibits the potential hardening in the spectrum above ~6 keV that appears to continue up to ~30 keV, beyond which the spectrum turns over. There is no obvious iron line in the XIS spectra.

Extended Data Figure 4 Suzaku lag–energy spectrum.

The lag–energy spectrum of the Suzaku observation taken 10 days before the XMM-Newton observation. An iron K lag at the same frequency and same energy as the XMM-Newton iron K lag is detected in the data. This confirms the presence of an iron K lag with an amplitude of ~120 s in two separate observations, taken with different instruments and analysed using different techniques. Error bars are at the 1σ level.

Extended Data Figure 5 Statistical significance of the XMM-Newton lag–energy spectrum.

a, The observed lag–energy spectrum (black points; same as Fig. 1 except the zero-point lag has not been shifted) compared with the 1σ distribution of 10,000 simulated Monte Carlo light curve pairs in each energy bin with zero lag (blue shaded region). This plot shows that the observed 0.3–10 keV lag energy is inconsistent with zero lag at >99% confidence. See Methods for details of the simulations. b, c, The best Gaussian (b) and asymmetric diskline (c) model fits (red lines) to the 3–10 keV lag–energy spectrum. We compute the Bayes’ factor between the models and find that the diskline is preferred at >98% confidence.

Extended Data Figure 6 Swift J1644+57 lag–energy spectrum compared with Seyfert galaxies.

Overlay of the reverberation signatures in nearby variable Seyfert galaxies compared with Swift J1644+57 (black diamonds). We show IRAS 13224−3809 in green, 1H 0707−495 in blue and Swift J2127.4+5654 in purple. The y axis shows arbitrary time lag units that have been scaled so that all of the sources have the same iron K lag amplitude. We show the scaled lag because all of the sources have different black hole masses (see Fig. 3). There is similarity in the broadband shape, although the iron K line profile of Swift J1644+57 appears to be slightly narrower and peaks at higher energies.

Extended Data Figure 7 Flux–energy and lag–energy spectra of reflection in super-Eddington flow.

Flux–energy (left) and lag–energy (right) spectra derived from our toy model for iron xxvi (rest-frame 6.97 keV) Kα line reverberation from an outflowing funnel wall. The results for terminal velocities of 0, 0.1c, 0.2c, 0.3c, 0.4c, 0.5c and 0.6c are shown (from left to right in peak energies).

Extended Data Table 1 Observations used in this analysis
Extended Data Table 2 Results of the fits to the lag–energy spectrum
Extended Data Table 3 Lags and masses of Seyfert galaxies with reverberation detections

PowerPoint slides

Source data

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kara, E., Miller, J., Reynolds, C. et al. Relativistic reverberation in the accretion flow of a tidal disruption event. Nature 535, 388–390 (2016).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing