Spectral-Timing Evolution of the Black Hole X-ray Binary LMC X-3: An XPoSat View

arXiv:2611.14223 · astro-ph.HE · Submitted 2026-11-10 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: We're looking at the paper: "Spectral-Timing Evolution of the Black Hole X-ray Binary LMC X-3".

Title and authors: Vera: The paper itself is titled "Spectral–Timing Evolution of the Black Hole X-ray Binary LMC X-three: An XPoSat View," and it comes from Sayantan Bhattacharya, Aman Kaushik, and Sudip Bhattacharyya <ref:2611.14223#pg1,Spectral–Timing Evolution of the Black Hole X-ray Binary LMC X>.

Jocelyn: It’s a very focused study because it uses this specific source, LMC X-three as a benchmark for accretion disk evolution across different spectral states <ref:2611.14223#pg1,accretion disk evolution across different spectral states>.

Subrahmanyan: When you look at the authors’ focus on this specific system, it helps narrow down the physics we need to consider when we try to build general models for how black holes feed on matter.

Vera: It shows that even in a persistent source like LMC X-three there is a clear transition between different accretion modes that we can measure directly with these new tools <ref:2611.14223#pg1>.

Jocelyn: It’s about using this data to constrain the physical processes happening very close to the black hole, where things get extremely hot and dense.

Subrahmanyan: Understanding that transition is vital because it helps us understand the physics governing matter collapse and how energy is released in these extreme environments.

The paper's summary: Vera: The main takeaway from this paper is that LMC X-three went through a significant change between two thousand twenty-four and two thousand twenty-five moving from a harder and more variable state to a brighter, softer, and less variable state <ref:2611.14223#pg1,harder and more variable state to a brighter, softer, and less variable>.

Jocelyn: They use the combined spectral fits to show that the inner disk temperature increased from about zero point eight four keV in two thousand twenty-four up to about one point one seven keV in two thousand twenty-five which happened alongside a nearly fivefold increase in the total X-ray flux <ref:2611.14223#pg3>.

Subrahmanyan: That flux increase coupled with the temperature rise suggests that the accretion disk became much hotter and more luminous during that later period, which is a direct indicator of how the mass accretion rate changed.

Vera: The timing analysis backs this up because they see enhanced low-frequency variability in two thousand twenty-four while the variability is significantly suppressed in two thousand twenty-five which matches what you expect from a softer state <ref:2611.14223#pg1>.

Jocelyn: They also confirmed that the spectral fitting was best done with an absorbed multicolour disk blackbody model, and adding extra Comptonization didn't really help improve those fits statistically.

Subrahmanyan: That confirmation is important because it tells us that for this specific evolution, the thermal disk alone is doing most of the heavy lifting in describing the spectrum.

The paper's improvements: Vera: The authors propose a few key things to improve how we study these systems. First, they suggest an automated spectral-timing state classification system.

Jocelyn: That AI system would analyze the hardness intensity and hardness rms diagrams across multiple epochs to automatically sort the source into hard or soft states without manual interpretation.

Subrahmanyan: If you can automate that sorting, it gives us a way to quickly map out the phase space of accretion states for black holes, which is crucial for testing our theories on how these flows behave.

Vera: They also suggest a system for predicting the evolution of fractional rms variability based on the trends in spectral hardness they see.

Jocelyn: That would use that established positive correlation between spectral hardness and rapid X-ray variability to predict how much noisier or smoother the source will be in the next observation cycle.

Subrahmanyan: That’s a practical tool, because if we can predict the variability suppression, we can better constrain the physical geometry of the inner flow during those softer phases.

Conclusion: Vera: So to wrap up, this paper on "Spectral–Timing Evolution of the Black Hole X-ray Binary LMC X-three: An XPoSat View" shows a clear path from a harder, more variable state in two thousand twenty-four to a brighter, softer one in two thousand twenty-five <ref:2611.14223#pg1,Spectral–Timing Evolution of the Black Hole X-ray Binary LMC X>.

Jocelyn: The core finding is that the thermal disk gets hotter and brighter, leading to suppressed rapid variability during that later epoch.

Subrahmanyan: This confirms the general picture of evolution toward a disk-dominated state, and it’s consistent with how we expect accretion flows to behave when they are dominated by thermal emission.

Vera: The measurements of the inner disk radius are consistent with the accretion disk extending close to the innermost stable circular orbit during that thermally dominated state, which is pretty tight.

Jocelyn: It’s a good result because it shows that XSPECT can successfully reproduce those canonical relationships between spectral hardness, rapid variability, and thermal disk evolution.

Subrahmanyan: Overall, this work validates the use of simultaneous spectral and timing diagnostics for monitoring bright Galactic black hole binaries over long timescales.

Vera: We're looking forward to how these observations help us refine our models of accretion physics in the next set of papers.

Sayantan Bhattacharya, Aman Kaushik, Sudip Bhattacharyya

Department of Astronomy and Astrophysics, Tata Institute of Fundamental Research

astro-ph.HE

Submitted: 2026-11-10

Updated: 2026-11-10

Comments: 12 pages, 5 figures, 2 tables. Published in ApJ

Journal ref: Sayantan Bhattacharya et al 2026 ApJ,1010,42

DOI: 10.3847/1538-4357/aea153

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 83/100

The gist: LMC X–3 exhibits significant changes in both its spectral and timing properties between two observing campaigns…

Key concepts

Spectral Evolution
This refers to how the X-ray spectrum changes over time. The study found that the inner accretion disk became hotter, increasing its temperature from about 0.84 keV in 2024 to 1.17 keV in 2025, which is a key indicator of the source's physical state.
Timing Analysis
This involves analyzing the variability of X-ray flux over time using power density spectra (PDS). The analysis showed that variability was significantly enhanced at low frequencies in 2024 but suppressed in 2025, reflecting a change in the source's underlying physical processes.
Hardness-RMS Diagram
This diagram plots X-ray hardness against fractional root mean square (rms) variability. The findings reveal a positive correlation: harder observations show larger variability, while softer observations show reduced variability, linking spectral state to timing behavior.
Disk-Dominated State
This describes an accretion state where the emission is primarily dominated by the thermal radiation from an optically thick accretion disk. LMC X–3 evolved toward this state in 2025, meaning it became brighter and softer as the disk's contribution grew.

Terminology

Summary

LMC X–3 exhibits significant changes in both its spectral and timing properties between two observing campaigns, indicating a transition from a relatively harder and more variable state to a brighter, softer, and less variable state The spectral and timing diagnostics provide a self-consistent picture of accretion-state evolution in LMC X–3

Spectral Evolution

The combined XSPECT spectra obtained during the 2024 and 2025 campaigns were fitted using an absorbed multicolour disk blackbody model The spectral fits indicate a significant increase in the inner disk temperature between 2024 and 2025, accompanied by an increase in X-ray flux by nearly a factor of five Both epochs are adequately described by the absorbed multicolour disk blackbody model tbabs×diskbb, while the addition of an extra Comptonization component does not produce a statistically significant improvement to the spectral fits The increase in disk temperature from kTin ≈ 0.84 keV in 2024 to 1. p. 2

Where the paper says it

The instrument offers an energy resolution of approximately 180 eV at 6 keV and a fast readout time of ∼ 1 ms, enabling timing studies of bright X-ray sources while minimizing pile-up effects [3, 11, 15]. Observations LMC X–3 was observed by XSPECT during two monitoring campaigns in 2024 and 2025.

p. 2
Where the paper says it

The instrument offers an energy resolution of approximately 180 eV at 6 keV and a fast readout time of ∼ 1 ms, enabling timing studies of bright X-ray sources while minimizing pile-up effects [3, 11, 15]. Observations LMC X–3 was observed by XSPECT during two monitoring campaigns in 2024 and 2025.

p. 2
Where the paper says it

The instrument offers an energy resolution of approximately 180 eV at 6 keV and a fast readout time of ∼ 1 ms, enabling timing studies of bright X-ray sources while minimizing pile-up effects [3, 11, 15]. Observations LMC X–3 was observed by XSPECT during two monitoring campaigns in 2024 and 2025.

17 keV in 2025, accompanied by an approximately fivefold increase in X-ray flux, indicates that the accretion disk became both hotter and more luminous during the later observations

Timing Analysis

The broad-band power density spectra (PDS) derived from the 2024 and 2025 observations are shown in Figure 4 The 2024 observations exhibit significantly enhanced variability power over a broad frequency range compared to 2025 Excess low-frequency variability is particularly evident below ∼ 0.05 Hz The suppression of broad-band variability during 2025 is consistent with the softer spectral state inferred from the HID and HRD analyses No statistically significant quasi-periodic oscillations are detected in either epoch

Hardness Diagnostics

The hardness–intensity diagrams (HIDs) for the two epochs are shown in Figure 2 The 2024 observations span a wide range of hardness values and intensities, indicating substantial spectral evolution during the observing campaign In contrast, the 2025 observations are concentrated at lower hardness ratios and higher count rates, consistent with a thermally dominated soft state The time evolution encoded in the color scale demonstrates a gradual migration of the source through the HID during the 2024 observations, while the 2025 observations remain comparatively stable

Variability Metrics

The hardness–rms diagram demonstrates that harder observations exhibit larger fractional rms variability systematically, whereas softer observations show substantially reduced variability This positive correlation between spectral hardness and rapid X-ray variability is a well-established property of black hole X-ray binaries As the thermal disk contributes an increasing fraction of the observed X-ray emission, the intrinsically stronger variability associated with the hot inner flow becomes progressively diluted, leading to lower fractional rms amplitudes The power density spectra support this interpretation by showing enhanced low-frequency broadband variability during the harder 2024 observations and a significant suppression of variability during the softer 2025 epoch

Conclusion

The spectral and timing diagnostics present a consistent picture of LMC X–3 evolving toward a disk-dominated accretion state characterized by increasing thermal emission and suppressed rapid variability These results establish XSPECT as a capable facility for future long-term monitoring of bright Galactic X-ray binaries and for investigating accretion-state evolution using simultaneous spectral and timing diagnostics The observed behavior is fully consistent with this spectral evolution The source evolves from a relatively harder, more variable, and fainter state during the 2024 observations to a brighter, softer, and more thermally dominated state during the 2025 campaign The hardness–intensity diagram shows a clear separation between the two observing epochs The source occupies both relatively hard and intermediate spectral states No statistically significant quasi-periodic oscillations or inter-band time lags were detected in either campaign The inferred radius is consistent with the accretion disk extending close to the innermost stable circular orbit during the thermally dominated state These results are in good agreement with more than two decades of observations of LMC X–3 obtained with missions such as RXTE, XMM–Newton, and NICER The present analysis demonstrates that the instrument successfully reproduces the canonical relationships between spectral hardness, rapid X-ray variability, and thermal disk evolution The results are in good agreement with the established phenomenology of black hole X-ray binaries The observed timing behaviour is fully consistent with this spectral evolution The power density spectra support this interpretation by showing enhanced low-frequency broadband variability during the harder 2024 observations and a significant suppression of variability during the softer 2025 epoch No statistically significant quasi-periodic oscillations or inter-band time lags were detected in either campaign The spectral evolution independently supports the same physical picture Both epochs are adequately described by the absorbed multicolour disk blackbody model tbabs×diskbb, while the addition of an extra Comptonization component does not produce a statistically significant improvement to the spectral fits The increase in inner disk temperature from kTin ≈ 0.84 keV in 2024 to 1.17 keV in 2025, accompanied by an approximately fivefold increase in X-ray flux, indicates that the accretion disk became both hotter and more luminous during the later observations These values are consistent with the range of disk temperatures previously reported for LMC X–3 during its thermally dominated state The fitted diskbb normalization provides a useful consistency check on the inferred accretion geometry Assuming a source distance of ∼ 50 kpc and a binary inclination of ∼ 67◦, the bestfitting normalization for the 2025 spectrum corresponds to an apparent inner disk radius of approximately 43 km Applying the standard corrections for spectral hardening and the stressfree inner boundary condition yields a physical radius of approximately 50 km For a black hole mass of ∼ 7 M⊙ and a dimensionless spin parameter of a∗ ≈ 0. p. 9

Where the paper says it

Assuming a source distance of ∼ 50 kpc and a binary inclination of ∼ 67◦ , the bestfitting normalization for the 2025 spectrum corresponds to an apparent inner disk radius of approximately 43 km. Applying the standard corrections for spectral hardening and the stressfree inner boundary condition [8] yields a physical radius of approximately 50 km.

p. 9
Where the paper says it

Assuming a source distance of ∼ 50 kpc and a binary inclination of ∼ 67◦ , the bestfitting normalization for the 2025 spectrum corresponds to an apparent inner disk radius of approximately 43 km. Applying the standard corrections for spectral hardening and the stressfree inner boundary condition [8] yields a physical radius of approximately 50 km.

25, the expected radius of the innermost stable circular orbit is approximately 53 km The present analysis therefore does not provide an independent measurement of the ISCO location, but rather demonstrates that the XSPECT observations are fully consistent with the established accretion geometry of the source The spectral and timing diagnostics present a consistent picture of LMC X–3 evolving toward a disk-dominated accretion state characterized by increasing thermal emission and suppressed rapid variability These results are in good agreement with the established phenomenology of black hole X-ray binaries The observed timing behaviour is fully consistent with this spectral evolution The hardness–rms diagram reveals that observations with larger hardness ratios exhibit systematically stronger fractional rms variability, whereas softer observations are characterized by substantially weaker variability This positive correlation between spectral hardness and rapid X-ray variability is a well-established property of black hole X-ray binaries and is generally interpreted as a consequence of the changing relative contributions of the optically thick accretion disk and the variable hard X-ray emitting region As the thermal disk contributes an increasing fraction of the observed X-ray emission, the intrinsically stronger variability associated with the hot inner flow becomes progressively diluted, leading to lower fractional rms amplitudes The power density spectra support this interpretation by showing enhanced low-frequency broadband variability during the harder 2024 observations and a significant suppression of variability during the softer 2025 epoch No statistically significant quasi-periodic oscillations or inter-band time lags were detected in either campaign The spectral evolution independently supports the same physical picture Both epochs are adequately described by the absorbed multicolour disk blackbody model tbabs×diskbb, while the addition of an extra Comptonization component does not produce a statistically significant improvement to the spectral fits The increase in inner disk temperature from kTin ≈ 0. p. 8
Where the paper says it

This positive correlation between spectral hardness and rapid X-ray variability is a well-established property of black hole X-ray binaries and is generally interpreted as a consequence of the changing relative contributions of the optically thick accretion disk and the variable hard X-ray emitting region [2]. As the thermal disk contributes an increasing fraction of the observed X-ray emission, the intrinsically stronger variability associated with the hot inner flow becomes progressively diluted, leading to lower fractional rms amplitudes[5].

84 keV in 2024 to 1.17 keV in 2025, accompanied by an approximately fivefold increase in X-ray flux, indicates that the accretion disk became both hotter and more luminous during the later observations These values are consistent with the range of disk temperatures previously reported for LMC X–3 during its thermally dominated state The fitted diskbb normalization provides a useful consistency check on the inferred accretion geometry Assuming a source distance of ∼ 50 kpc and a binary inclination of ∼ 67◦, the bestfitting normalization for the 2025 spectrum corresponds to an apparent inner disk radius of approximately 43 km Applying the standard corrections for spectral hardening and the stressfree inner boundary condition yields a physical radius of approximately 50 km For a black hole mass of ∼ 7 M⊙ and a dimensionless spin parameter of a∗ ≈ 0.25, the expected radius of the innermost stable circular orbit is approximately 53 km The present analysis therefore does not provide an independent measurement of the ISCO location, but rather demonstrates that the XSPECT observations are fully consistent with the established accretion geometry of the source

Key Findings Summary

The principal results are summarized as follows:

[2611.14223] The broad-band light curves and hardness evolution reveal significant long-term spectral evolution, with the 2024 observations sampling a harder and more variable state, while the 2025 observations correspond to a brighter, softer, and more stable accretion state. p. 9

Where the paper says it

By combining hardness diagnostics, timing analysis, and X-ray spectral modeling, we characterize the long-term evolution of the source and demonstrate the capability of XSPECT to investigate accretion-state evolution in bright black hole X-ray binaries. Our principal results are summarized as follows: • The broad-band light curves and hardness evolution reveal significant long-term spectral evolution, with the 2024 observations sampling a harder and more variable state, while the 2025 observations correspond to a brighter, softer, and more stable accretion state.

[13] The hardness–intensity diagram shows a clear separation between the two observing epochs, while the 2024 observations themselves trace a gradual evolution from harder to softer states over the course of the monitoring campaign. p. 9

Where the paper says it

Our principal results are summarized as follows: • The broad-band light curves and hardness evolution reveal significant long-term spectral evolution, with the 2024 observations sampling a harder and more variable state, while the 2025 observations correspond to a brighter, softer, and more stable accretion state. • The hardness–intensity diagram shows a clear separation between the two observing epochs, while the 2024 observations themselves trace a gradual evolution from harder to softer states over the course of the monitoring campaign.

[13] The hardness–rms diagram demonstrates that harder observations exhibit larger fractional rms variability systematically, whereas the softer observations show substantially reduced variability, consistent with an increasing contribution from a stable thermal accretion disk. p. 8

Where the paper says it

The hardness– rms diagram reveals that observations with larger hardness ratios exhibit systematically stronger fractional rms variability, whereas softer observations are characterized by substantially weaker variability. This positive correlation between spectral hardness and rapid X-ray variability is a well-established property of black hole X-ray binaries and is generally interpreted as a consequence of the changing relative contributions of the optically thick accretion disk and the variable hard X-ray emitting region [2].

[10] Broad-band power density spectra reveal enhanced low-frequency variability during the 2024 observations, while the variability is significantly suppressed in 2025. p. 7

Where the paper says it

10 3 10 2 10 1 Frequency (Hz) 10 2 10 1 f P(f) [F r a c tio n al r m s 2 ] 2024 2025 Figure 4: Broad-band power density spectra of LMC X–3 during 2024 and 2025 XSPECT/XpoSat observations. The harder 2024 state exhibits enhanced low-frequency variability relative to the softer 2025 observations.

[9] Spectral fitting with an absorbed multicolor disk blackbody model indicates an increase in the inner disk temperature from kTin ≈ 0. p. 8

Where the paper says it

8 1 2 3 4 5 7 Energy (keV) 0. 2 Ratio Figure 5: XSPECT/XpoSat spectra of LMC X–3 obtained during the 2024 (soft only) and 2025 observing campaigns together with the best-fitting absorbed multicolour disk blackbody model (tbabs × diskbb).

84 keV in 2024 to ≈ 1. p. 2
Where the paper says it

The instrument offers an energy resolution of approximately 180 eV at 6 keV and a fast readout time of ∼ 1 ms, enabling timing studies of bright X-ray sources while minimizing pile-up effects [3, 11, 15]. Observations LMC X–3 was observed by XSPECT during two monitoring campaigns in 2024 and 2025.

17 keV in 2025, accompanied by an increase in the X-ray flux by nearly a factor of five. p. 8
Where the paper says it

84 keV in 2024 to 1. 17 keV in 2025, accompanied by an approximately fivefold increase in X-ray flux, indicates that the accretion disk became both hotter and more luminous during the later observations.

[13] The inferred inner disk radius is consistent with the accretion disk extending close to the innermost stable circular orbit during the thermally dominated state. p. 9

Where the paper says it

25, the expected radius of the innermost stable circular orbit (ISCO) is approximately 53 km[10, 13]. Although the diskbb normalization alone cannot provide a precise measurement of the inner disk radius because of uncertainties in the colour-correction factor, inclination, and relativistic effects, the inferred radius is nevertheless consistent with previous continuum-fitting studies indicating that the accretion disk in LMC X–3 remains close to the ISCO during thermally dominated states [13, 14].

The observed timing behaviour is fully consistent with this spectral evolution The spectral and timing diagnostics present a consistent picture of LMC X–3 evolving toward a disk-dominated accretion state characterized by increasing thermal emission and suppressed rapid variability These results are in good agreement with the established phenomenology of black hole X-ray binaries The instrument successfully reproduces the canonical relationships between spectral hardness, rapid X-ray variability, and thermal disk evolution The present analysis demonstrates that the instrument successfully reproduces the canonical relationships between spectral hardness, rapid X-ray variability, and thermal disk evolution

Instrument Capability

The X-ray Polarimeter Satellite (XPoSat) carries the X-ray Spectroscopy and Timing (XSPECT) payload, which provides simultaneous spectroscopic and timing observations in the 0.8–15 keV energy range XSPECT consists of an array of Swept Charge Devices (SCDs) arranged in four detector quadrants, providing fields of view of 2◦ × 2◦ and 3◦ × 3◦ The instrument offers an energy resolution of approximately 180 eV at 6 keV and a fast readout time of ∼ 1 ms, enabling timing studies of bright X-ray sources while minimizing pile-up effects

Data Sources

The XPoSat/XSPECT data analysed in this work are publicly available through the Indian Space Science Data Centre (ISSDC) The processed data products and analysis scripts used in this work are available from the corresponding author upon reasonable request The X-ray Polarimeter Satellite was launched on 2024 January 1 as India’s first dedicated X-ray polarimetry mission Data were downloaded from the Indian Space Science Data Centre (ISSDC) archive for all publicly available Level-2 XSPECT products corresponding to these observations The Level-2 event files, which already incorporate the recommended screening criteria and Good Time Intervals (GTIs), were adopted for all subsequent analysis The archival Level-2 event files were filtered from flares using xspfilterflares command The observations consisted of multiple individual exposures in each year, the Level-2 event files were merged using the xspevtmerge task to create epoch-averaged event files for spectral analysis Source spectra were generated using the standard XSPECT spectral extraction command (xspspecgen) and the corresponding response matrix files (RMFs) (provided from the XSPECT archive) and ancillary response files (ARFs) generated by the pipeline Light curves were extracted directly from the Level-2 event files in multiple energy bands using xsplcgen command Broad (0.8–10.0 keV), soft (0. p. 4

Where the paper says it

To investigate spectral evolution, broad (0. 0 keV), soft (0.

8–2.0 keV), and hard (2. p. 4
Where the paper says it

0 keV), soft (0. 0 keV), and hard (2.

0–10.0 keV) light curves (500s time bin) were produced using the XSPECT pipeline Hardness ratios were defined as the ratio of the hard-band to soft-band count rates Hardness–intensity and hardness–rms diagrams were constructed using these light curves after removing statistically significant outliers through sigma-clipping Power density spectra were computed from the broadband light curves and normalized following the fractional rms prescription of Belloni et al. p. 4
Where the paper says it

Hardness ratios were defined as the ratio of the hard-band to soft-band count rates. Hardness–intensity and hardness–rms diagrams were constructed using these light curves after removing statistically significant outliers through sigma-clipping.

[2] Hardness-dependent variability was further quantified using hardness–rms diagrams We additionally searched for inter-band delays between the soft and hard energy bands using cross-correlation of light curves binned at 10 milliseconds The spectra were grouped to ensure sufficient counts per bin for χ2 minimization The uncertainties quoted in this work correspond to the 68 per cent confidence interval for a single interesting parameter The XPoSat/XSPECT data analysed in this work are publicly available through the Indian Space Science Data Centre (ISSDC) The processed data products and analysis scripts used in this work are available from the corresponding author upon reasonable request The usage of AI has been limited to editing and rephrasing the document; AI has not been used for any creative purposes

References

[1] Arnaud, K., Dorman, B., & Gordon, C. 1999, Astrophysics Source Code Library, ascl

[2] Belloni, T., Psaltis, D., & van der Klis, M. 2002, The Astrophysical Journal, 572, 392

[3] Chatterjee, R., Vadodariya, K., Vatedka, R., et al. 2025, Journal of Astronomical Telescopes, Instruments, and Systems

[4] Cowley, A., Crampton, D., Hutchings, J., et al. 1984, Astrophysical Journal, Part 1 (ISSN 0004-637X), vol. 286, Nov. 1, 1984

[5] Done, C., Gierliński, M., & Kubota, A. 2007, The Astronomy and Astrophysics Review, 15

[6] Dunn, R., Fender, R., Körding, E., Belloni, T. p. 4

Where the paper says it

Unless otherwise noted, all uncertainties quoted in this work correspond to the 68 per cent confidence interval for a single interesting parameter. For timing studies, power density spectra were computed from the broadband light curves and normalized following the fractional rms prescription of Belloni et al.

, & Cabanac, C. 2010, Monthly Notices of the Royal Astronomical Society

[7] Kubota, A., Done, C., Davis, S. W., et al. 2010, The Astrophysical Journal

[8] Kubota, A., Tanaka, Y., Makishima, K., et al. 1998, PASJ

[9] Nowak, M., Wilms, J., Heindl, W., et al. 2001, Monthly Notices of the Royal Astronomical Society

[10] Orosz, J. A., Steiner, J. F., McClintock, J. E., et al. 2014, The Astrophysical Journal

[11] Prakash, V. S., Agrawal, V. K., Chatterjee, R., et al. 2025, Monthly Notices of the Royal Astronomical Society

[12] Remillard, R. A., & McClintock, J. E. 2006, Annu. Rev. Astron. Astrophys., 44

[13] Steiner, J. F., McClintock…

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Automated spectral-timing state classification: An improved AI system can automatically classify black hole X-ray binary accretion states by analyzing combined hardness intensity diagrams and hardness rms diagrams to distinguish between hard and soft states across multiple epochs.

  2. Long-term variability trend prediction: The system can predict the evolution of fractional rms variability based on spectral hardness trends, utilizing the established positive correlation between spectral hardness and rapid X-ray variability found in the data.

  3. Disk temperature calibration: The AI can estimate changes in inner disk temperature by fitting absorbed multicolour disk blackbody models to combined spectra, specifically tracking the increase from 0.84 keV in 2024 to approximately 1.17 keV in 2025.

  4. Accretion rate estimation: By correlating the increase in X-ray flux (nearly a factor of five) with the temperature increase, the system can estimate changes in mass accretion rate as it evolves toward a more thermally dominated state during 2025.

Abstract

LMC X-3 is a persistent black hole high-mass X-ray binary and serves as an important laboratory for studying accretion disk evolution across different spectral states. We present a spectral-timing analysis of XPoSat/XSPECT observations of LMC X--3 obtained during 2024 and 2025. Using broadband X-ray light curves, hardness diagnostics, power spectral analysis, and spectral modeling, we investigate the long-term evolution of the source across multiple epochs. LMC X-3 exhibits significant changes in both its spectral and timing properties between the two observing campaigns. Hardness-intensity and hardness-rms diagrams reveal a clear separation between the 2024 and 2025 observations and even within the 2024 observations. Overall, it indicates a transition from a relatively harder and more variable state to a brighter, softer, and less variable state. Power spectral analysis shows enhanced broadband variability during the 2024 observations, while the 2025 observations are characterized by weaker variability. Spectral modeling of the combined XSPECT observations indicates an increase in the thermal disk temperature from 0.8 to 1.17 keV, accompanied by a ~5 times increase in X-ray flux between the two epochs. The observed reduction in fractional rms variability with increasing source brightness and decreasing spectral hardness is consistent with more contribution from a stable thermal accretion disk. Taken together, the spectral and timing diagnostics provide a self-consistent picture of accretion-state evolution in LMC X-3, demonstrating the capability of XSPECT to probe long-term variability and accretion processes in bright black hole X-ray binaries.

Related papers