An intrinsic decline of accretion activity in GRS 1915+105

Technical Summary: An intrinsic decline of accretion activity in GRS 1915+105

Problem Statement
The canonical microquasar GRS 1915+105 has exhibited unprecedented multiwavelength changes since 2018, entering a prolonged low-flux, heavily obscured X-ray state punctuated by intense flaring in 2023–2024. Recent deep radio and soft X-ray non-detections have left the nature of the source's current core activity ambiguous. The central problem addressed is whether the source's disappearance is due to an enhanced, deeply obscuring circumnuclear column (a "cocoon") or an intrinsic decline in accretion activity. These two scenarios are currently degenerate, as both can suppress observed flux.

Methodology
The authors present new multiwavelength constraints to break this degeneracy:

1. NuSTAR X-ray Observations: A pointed observation on November 17, 2025, provided a deep non-detection in the 3–78 keV band. The authors utilized Bayesian X-ray Analysis (BXA) with nested sampling (MultiNest) to derive rigorous flux upper limits ($F_{3-78} < 1.3 \times 10^{-13}$ erg s$^{-1}$ cm$^{-2}$), corresponding to a luminosity limit of $L_{3-78} < 1.4 \times 10^{33}$ erg s$^{-1}$. This limit is $10^4$–$10^6$ times deeper than historical detections.
2. SPHEREx Infrared Spectrophotometry: Data from the SPHEREx mission (epochs in late 2025 and early 2026) were analyzed to construct the source's spectral energy distribution (SED) in the 0.75–5.0 $\mu$m range. The authors employed forced photometry to mitigate contamination from the crowded infrared field, acknowledging that residual contamination may exist.
3. Modeling and Comparison: The authors tested the "enhanced obscuration" hypothesis by fitting the NuSTAR non-detection with physically self-consistent AGN obscuration models (UXCLUMPY, XSKIRTOR, and BNsphere). They calculated the required scattering fractions and column densities to explain the X-ray and radio non-detections. Furthermore, they estimated the thermal reprocessing luminosity expected from a Compton-thick cocoon absorbing the source's historical accretion power and compared it to the observed infrared flux. The source's position was also mapped onto the mid-infrared–X-ray ($L_{\text{MIR}}$–$L_X$) plane relative to historical data and other black hole binaries.

Key Results

• X-ray Constraints: The NuSTAR non-detection requires that if the source is intrinsically active, any obscuring cocoon must be Compton-thick ($N_H \gtrsim 2 \times 10^{25}$ cm$^{-2}$) and fully sky-covering ($4\pi$). Crucially, to suppress the flux to observed levels, the scattering fraction ($f_{\text{scatt}}$) must be less than $\sim 10^{-4}$. Standard models of Compton-thick absorbers typically exhibit scattering fractions of a few percent; achieving such low scattering requires an implausibly specific geometry.
• Infrared Constraints: If the source were intrinsically active at historical levels ($L_{\text{acc}} \sim 0.05 L_{\text{Edd}}$) and fully obscured by a Compton-thick shell, the absorbed energy would be reprocessed into thermal infrared emission. For a shell radius of $10^{13}$ cm, the predicted flux at 5 $\mu$m would be $\sim 70$ mJy (reddened), exceeding the SPHEREx detection limits by nearly two orders of magnitude. Even for larger shells, the required mass to sustain such a column becomes unphysically large (comparable to the entire binary system mass).
• SED and Plane Analysis: The broadband SED constructed from NuSTAR, SPHEREx, and radio data shows the source is significantly fainter across the electromagnetic spectrum compared to its historical states. On the $L_{\text{MIR}}$–$L_X$ plane, the source now occupies the regime consistent with quiescent black hole X-ray binaries, rather than the obscured or flaring states seen previously.
• Radio Non-detection: The authors note that an ionized, Compton-thick cocoon would be optically thick to radio waves via free-free absorption, consistent with the deep radio non-detections. However, this does not resolve the conflict with the lack of reprocessed infrared emission.

Conclusion and Significance
The paper argues that the "enhanced obscuration" scenario is inconsistent with the lack of reprocessed thermal emission in the infrared and the extreme geometric constraints required to suppress scattered X-rays. Instead, the data are consistent with an intrinsic weakening of accretion activity by 2–4 orders of magnitude (or more) occurring around November 2025 or earlier.

The authors propose a physical mechanism wherein the violent outflows and flaring activity observed during 2023–2024 progressively expelled the fueling material from the inner accretion disc, leading to a quenching of the source. This interpretation suggests GRS 1915+105 is transitioning into a new quiescent phase. The significance of this work lies in providing the first clear evidence of such dramatic secular accretion changes on human timescales in a microquasar. The authors suggest that continued monitoring, particularly with higher-frequency radio observations and sensitive infrared/sub-mm facilities (like ALMA and JWST), is necessary to confirm the depletion of the inner disc and characterize any remaining gaseous cocoon.