Research ArticleASTROPHYSICS

Accretion-induced variability links young stellar objects, white dwarfs, and black holes

See allHide authors and affiliations

Science Advances  09 Oct 2015:
Vol. 1, no. 9, e1500686
DOI: 10.1126/sciadv.1500686


  • Fig. 1 Kepler/K2 light curve of V866 Sco.

    Light curve of the YSO object V866 Sco (EPIC205249328) obtained with the Kepler/K2 mission during campaign 2 in long cadence mode (29.4 min). The light curve was extracted from the target pixel files provided by MAST by manually defining large target and background masks.

  • Fig. 2 Power spectrum and rms-flux relation of the accreting YSO V866 Sco.

    We show the obtained power spectrum of the accreting YSO star V866 Sco (AS 205A or EPIC205249328). The rms-normalized (42) power spectrum was estimated using Kepler data from campaign 2, comprising over 78 days of continuous observations with 29.4-min cadence. We mark the measured characteristic bend frequency with the vertical gray solid line, and 1σ limits with the gray-shaded region (19). The inset shows the linear rms-flux relation from the light curve in units of 105 electrons/s (black data points), computed on 2.5-hour time scales. The gray line is a linear fit to the data. We are able to measure characteristic bend frequencies for the other five YSOs in the Kepler sample (which include EPIC204181799, EPIC204395393, EPIC204830786, EPIC204908189, and EPIC205110000), as well as linear rms-flux relations, on a wide range of time scales.

  • Fig. 3 Confidence contours for the mass, radius, and mass accretion rate indices.

    We show the confidence contours for the dependency of the characteristic bend frequency on the accretor mass, radius, and mass accretion rate indices. We have fitted the model, log νb = A log R + B log M + C log Embedded Image + D, where νb represents the measured characteristic bend frequencies, and R, M, and Embedded Image are the radii, masses, and mass accretion rates, respectively (all using cgs units), to determine the best-fit values for A, B, C, and D. The obtained fit is good (X2 = 38.41 for 37 df) and consistent with the previously obtained fit (1), resulting in A = −2.07 ± 0.11, B = 0.043 ± 0.17, C = 0.95 ± 0.22, and D = −3.07 ± 2.61. The dark gray lines show the previously obtained fit using a BH-only sample (1), with 1σ errors marked with light gray regions. Given that the previous fit was solely based on BHs, a degeneracy between mass and radius existed, such that it was essentially fitting for νbRaMb and the finding that a + b ≈ −2. The contours refer to the fit including accreting WDs, AGN, and soft-state Galactic BHs and allowing the radius to be an additional free parameter. We thus show that the dominant parameter in the scaling law is a characteristic radius in the inner disc, and not the mass of the accretor as previously thought.

  • Fig. 4 Edge-on projection of our sample on the accretion variability plane.

    We show that the predicted characteristic bend frequencies, derived by inserting the observed masses, radii, and mass accretion rates into the best-fit relationship to the combined supermassive BHs, stellar-mass BHs, and accreting WDs, agree very well with the observed characteristic bend frequencies. If the predicted and observed bend frequencies are identical, then an object will exactly lie on the black line. We display supermassive BHs as filled green squares, stellar-mass BHs in their soft state as filled red squares, and accreting WDs as filled blue squares. We additionally show the position of the YSO V866 Sco with the filled magenta circle using the best estimate for its magnetospheric radius at 6.4 solar radii. The error on the predicted νb assumes strict upper and lower limits for the disc truncation radius, from 0.1 to 2 AU solar radii (30). We thus demonstrate that the variability plane of accreting systems extends from supermassive BHs all the way to YSOs.


  • Table 1 Samples and parameters used in this work.

    The table lists the objects used in this work. We include the adopted masses, radii, mass accretion rates, and break frequencies, νb. Where these values have been taken from the literature, we provide the relevant reference. Where these values have been estimated in this work, we refer to Materials and Methods (19). All systems, except for the YSO V866 Sco, have been used for the fit shown in Figs. 3 and 4.

    NameTypeLog(mass) [log(M)]Log(radius) [log(R)]Log(Embedded Image) [log(M/year)]log(νb) [log(Hz)]References
    Cyg X-1BH1.17−4.03−8.381.09(3, 22, 34, 35)
    GRS 1915+105BH1.18−4.02−7.061.87(3, 7, 23, 36)
    Mrk 335AGN7.342.14−1.06−3.76(21)
    H 0707-495AGN6.371.17−1.32−3.82(21)
    ESO 434-G40AGN6.301.10−1.45−4.06(21)
    NGC 3227AGN6.881.68−1.89−3.64(21)
    KUG 1031+398AGN6.601.40−0.93−3.44(21)
    NGC 4051AGN6.130.93−2.19−3.53(21)
    Mrk 766AGN6.571.37−1.67−3.70(21)
    NGC 4395AGN5.550.35−4.38−2.66(21)
    MCG 06.30.015AGN6.711.51−1.90−3.75(21)
    NGC 5506AGN7.462.26−1.75−3.78(21)
    NGC 6860AGN7.592.39−1.22−4.03(21)
    Akn 564AGN6.901.70−0.85−3.46(21)
    NGC 3516AGN7.502.30−1.46−5.70(21)
    NGC 3783AGN7.472.27−1.34−5.20(21)
    NGC 4151AGN7.121.92−2.02−5.90(21)
    Fairall 9AGN8.413.21−0.52−6.39(21)
    NGC 5548AGN7.642.44−0.92−6.22(21)
    NGC 7469AGN7.392.19−1.19−6.00(21)
    PKS 0558-504AGN8.483.280.99−5.17(21)
    IC 4329AAGN8.343.14−0.25−5.60(21)
    PG 0804761AGN8.843.640.17−6.02(1)
    NGC 3516AGN7.632.43−1.49−5.70(1)
    NGC 4258AGN7.592.39−3.61−7.65(1)
    KIC 8751494WD−0.10−2.00−8.00−2.98(10, 19)
    MV LyrWD−0.14−1.98−8.00−2.89(8, 19)
    BZ CamWD−0.26−1.92−8.00−3.23(19, 25)
    CM DelWD−0.32−1.90−8.00−3.10(19, 25)
    KR AurWD−0.23−1.93−8.00−3.18(19, 25)
    RW TriWD−0.26−1.92−8.00−2.94(19, 25)
    UU AqrWD−0.17−1.96−8.00−2.70(19, 25)
    V345 PavWD−0.12−1.99−8.00−3.09(19, 25)
    V866 ScoYSO0.130.80−6.14−5.81(30)

Navigate This Article