Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Comprehensive X-ray Spectral-timing Analysis of GRS 1915+105 Based on Insight-HXMT Observations

Xiao Chen, Weiping Liu, Wei Wang

TL;DR

We address the complex X-ray variability of GRS 1915+105 with a broadband spectral-timing analysis using the full Insight-HXMT dataset from 2017 to 2023. We identify a QPO frequency rising branch from $\sim$2 Hz to $\sim$6 Hz during the decay and a third flare in the obscured state with distinctive spectral–timing properties, along with sub-Hz QPOs at $\sim$0.01–0.2 Hz across flares. A comparison between QPOs above and below 1 Hz suggests different physical origins: $>1$ Hz QPOs likely arise from Lense–Thirring precession of the inner hot flow, while $<1$ Hz QPOs reflect magnetic perturbations driving a magnetically driven failed disk wind. Together, these results constrain the evolving accretion geometry and wind–corona coupling in GRS 1915+105 and demonstrate the value of long-term broadband monitoring for revealing transitional behavior.

Abstract

GRS 1915+105 has been well studied since its discovery, and is well-known for its complex light curve variability. Using the full currently available Insight-HXMT dataset from July 2017 to June 2023, we make a comprehensive spectral-timing analysis of this source and report four main findings. First, we uncover a QPO frequency rising branch between MJD 58206 and 58230, where the centroid frequency increases from $\sim$2 Hz to $\sim$6 Hz, consistent with a spectral state transition from the hard to intermediate state. This rising branch completes the full QPO frequency evolution cycle when combined with the subsequent frequency decay phase, and had been missed in prior NICER and Insight-HXMT studies. Second, we identify a previously unreported Flare 3 during the obscured state, which shows distinct spectral and timing properties compared to the earlier flares. Third, we detect sub-Hz QPOs (<1 Hz) in all three flares, specifically at $\sim$0.01 Hz in Flare 1 and $\sim$0.2 Hz in both Flares 2 and 3. In particular, the weak $\sim$0.2 Hz signals observed in Flare 3 indicate ongoing coronal activity despite strong obscuration. Finally, a comparison between QPOs above and below 1 Hz suggests distinct origins, with the former likely arising from Lense-Thirring precession of the inner hot flow and the latter from magnetic perturbations driving a failed disk wind. These findings offer new insights into the unique accretion geometry and variability behaviors of GRS 1915+105.

Comprehensive X-ray Spectral-timing Analysis of GRS 1915+105 Based on Insight-HXMT Observations

TL;DR

We address the complex X-ray variability of GRS 1915+105 with a broadband spectral-timing analysis using the full Insight-HXMT dataset from 2017 to 2023. We identify a QPO frequency rising branch from 2 Hz to 6 Hz during the decay and a third flare in the obscured state with distinctive spectral–timing properties, along with sub-Hz QPOs at 0.01–0.2 Hz across flares. A comparison between QPOs above and below 1 Hz suggests different physical origins: Hz QPOs likely arise from Lense–Thirring precession of the inner hot flow, while Hz QPOs reflect magnetic perturbations driving a magnetically driven failed disk wind. Together, these results constrain the evolving accretion geometry and wind–corona coupling in GRS 1915+105 and demonstrate the value of long-term broadband monitoring for revealing transitional behavior.

Abstract

GRS 1915+105 has been well studied since its discovery, and is well-known for its complex light curve variability. Using the full currently available Insight-HXMT dataset from July 2017 to June 2023, we make a comprehensive spectral-timing analysis of this source and report four main findings. First, we uncover a QPO frequency rising branch between MJD 58206 and 58230, where the centroid frequency increases from 2 Hz to 6 Hz, consistent with a spectral state transition from the hard to intermediate state. This rising branch completes the full QPO frequency evolution cycle when combined with the subsequent frequency decay phase, and had been missed in prior NICER and Insight-HXMT studies. Second, we identify a previously unreported Flare 3 during the obscured state, which shows distinct spectral and timing properties compared to the earlier flares. Third, we detect sub-Hz QPOs (<1 Hz) in all three flares, specifically at 0.01 Hz in Flare 1 and 0.2 Hz in both Flares 2 and 3. In particular, the weak 0.2 Hz signals observed in Flare 3 indicate ongoing coronal activity despite strong obscuration. Finally, a comparison between QPOs above and below 1 Hz suggests distinct origins, with the former likely arising from Lense-Thirring precession of the inner hot flow and the latter from magnetic perturbations driving a failed disk wind. These findings offer new insights into the unique accretion geometry and variability behaviors of GRS 1915+105.

Paper Structure

This paper contains 15 sections, 2 equations, 12 figures.

Table of Contents

  1. Introduction
  2. Observations and Data Reduction
  3. Results
  4. Timing analysis
  5. Light curve variability
  6. QPOs above 1 Hz
  7. QPOs below 1 Hz
  8. Spectral analysis
  9. Spectral modelling
  10. Spectral results
  11. Discussion
  12. The QPO evolution in the decay phase
  13. Suspected type-B QPOs
  14. Distinct features of Flare 3 and sub-Hz QPOs
  15. Conclusion

Figures (12)

  • Figure 1: Top panel: Light curves from LE (dark blue), ME (emerald green), HE (orange), MAXI (grey), and Swift/BAT (red) are shown. All data points are shown with lightgrey error bars. The MAXI data correspond to the 2--20 keV energy range with units of ph s$^{-1}$ cm$^{-2}$, and the BAT data cover the 15--50 keV range with units of count s$^{-1}$ cm$^{-2}$. However, these units are not explicitly shown in the plot to avoid clutter. The y-axis label only indicates the units for the three Insight-HXMT detectors. Different states are marked with different background colors, with Soft Peach Pink represent the Active State (or Soft State), Pale Peach represent the Decay Phase, while Soft Blue represent the Flares. The Obscured Faint State are displayed with no (or white) background. Bottom panel: Hardness ratio Evolution. The hardness ratio is calculated based on 3--10 keV / 1--3 keV rates in LE.
  • Figure 2: Top panel: Relation between the hardness ratio (3--10 keV to 1--3 keV) and LE (1--10 keV) count rate. Data from the Soft State, Decay Phase, Flare 1, Flare 2 and Flare 3 are shown in red, light tan, blue, green and purple, respectively, with the Flare 3 points connected in chronological order. The black points represent the obscured faint states. All data points are shown with error bars, although they are too small to be clearly visible in the plot. Bottom panel: The relation between two X-ray colors. Hardness is defined as in the top panel, while Hardness2 is defined as the ratio of 10--35 keV to 1--3 keV count rate. The gray dashed ellipses indicate the approximate regions of the three states defined by Belloni2000AA...355..271B.
  • Figure 3: From top to bottom, the panels show the evolution of QPO centroid frequency above 1 Hz, FWHM, fractional rms, and significance over time. Data points with significance $\geq$3 are plotted in blue circles (LE), green squares (ME), and orange triangles (HE), while those with significance $<$3 are shown in grey. All data points are shown with light grey error bars. In the top panel, the region enclosed by the red dashed lines highlights the rising phase of the QPO centroid frequency. While in the bottom panel, the horizontal red dashed line marks the 3 $\sigma$ threshold.
  • Figure 4: Representative LE power spectra for the three main evolutionary phases (a-c) and the three flares (d-f). QPOs with frequencies above 1 Hz are commonly detected during the decay phase (panel b) and occasionally during the obscured state (panel c), whereas QPOs with frequencies around 0.2 Hz appear in the second flare (panel e).
  • Figure 5: Top panel: The ME PDS for observation P010131000202 (MJD 58055). The residuals in black show the best fitted results, while residuals in red indicate the fitting results after subtracting the QPO component (shifted upward by 5 for clarity). A distinct peak is visible around 2 Hz. Bottom panel: The LE PDS for the same observation. The broad peak around 1 Hz could not be well fitted with a Lorentzian.
  • ...and 7 more figures