Black Hole Measurements May Be Systematically Wrong Due to Oversimplified Disk Models

For decades, astronomers have used the shape of iron emission lines to measure black hole spin—one of only two fundamental properties of these cosmic giants. But new research suggests these measurements may be systematically underestimating both spin rates and other critical parameters because of a flawed assumption: that accretion disks are perfectly flat.

A preprint analysis by Surgent and colleagues employs advanced general relativistic ray tracing to model how X-ray light behaves around black holes when disks have realistic three-dimensional structure. Their findings indicate that when thick, warped, or radiation-pressure-dominated disks are analyzed using standard flat-disk models, the resulting measurements can significantly underestimate black hole spin, corona height, and viewing angle.

Why This Matters Now

The timing of this research is crucial. XRISM (X-Ray Imaging and Spectroscopy Mission), launched by JAXA in 2023, represents a generational leap in spectroscopic resolution for X-ray astronomy. The authors specifically frame their analysis using XRISM's measurement uncertainties, suggesting that this new observatory's unprecedented precision could finally make these geometric subtleties detectable—or alternatively, could systematically bias our measurements if we continue using oversimplified models.

Black hole spin is not merely an academic curiosity. It encodes the accretion history of the black hole, affects jet launching mechanisms that influence galaxy evolution, and provides one of the few testable predictions of general relativity in the extreme-gravity regime. If our spin measurements have been systematically biased downward, it could reshape our understanding of black hole growth, feedback processes, and relativistic jet physics.

The Iron Line: A Cosmic Speedometer

The iron K-alpha line at 6.4 keV serves as an exquisite probe of spacetime geometry near black holes. When X-rays from the hot corona above the disk illuminate the cooler material below, iron atoms fluoresce, emitting characteristic photons. But these photons don't simply travel to our telescopes unchanged.

General relativity warps their journey. Gravitational redshift stretches their wavelengths. Doppler shifts from orbital motion around the black hole further distort them—material approaching us appears blue-shifted, while receding material appears red-shifted. The resulting "broadened" iron line encodes information about the innermost stable circular orbit (ISCO), which depends directly on black hole spin.

Previous analyses, dating back to foundational work by Laor (1991) and more recent refinements by Dauser et al. (2013) with the RELXILL models, have predominantly assumed razor-thin, flat disks extending down to the ISCO. This simplification makes calculations tractable but ignores well-established physics.

What Real Disks Look Like

Theoretical models of accretion disks, particularly the Shakura-Sunyaev framework developed in 1973, predict that disk thickness varies with radius. In regions where radiation pressure dominates—precisely the hot inner regions near supermassive black holes where the iron line originates—disks develop substantial vertical scale heights.

Surgent's team modeled several realistic geometries:

Constant aspect ratio disks, where thickness scales proportionally with radius—a common approximation in numerical simulations.

Radiation-pressure-dominated Shakura-Sunyaev disks, which naturally puff up in their inner regions where temperatures and radiation pressure peak.

Expanded inner disks with non-negligible scale height specifically in the crucial innermost zones.

Warped disks, where the disk plane twists with radius due to Lense-Thirring precession—a general relativistic effect where rotating black holes drag spacetime, forcing misaligned disks to precess.

The Systematic Bias

The most concerning finding: when these realistic disk geometries are fitted with standard flat-disk models, the inferred parameters are systematically biased. Specifically, black hole spin is underestimated, as are corona height and inclination angle.

This happens because disk thickness changes how X-rays from the corona illuminate the disk and how reflected light escapes to reach observers. A thick disk geometrically blocks some sight lines and shadows certain regions, altering the observed line profile. Light bending—the relativistic focusing of photons toward the black hole and then back out along the disk—interacts differently with three-dimensional structure.

For warped disks, the situation is worse: the team found these geometries "could not be fit" with flat-disk approximations at all. This suggests that some observed systems might be fundamentally mischaracterized if warp is present but unaccounted for.

What Previous Work Missed

While the effects of disk geometry on spectral features have been explored before—notably by Vincent et al. (2016) and Fragile et al. (2007) in the context of simulations—most observational fitting tools still rely on flat-disk assumptions. The gap between theoretical understanding and practical analysis tools has persisted largely because ray tracing through complex geometries is computationally expensive, and observational data quality hasn't historically warranted the added complexity.

XRISM changes this calculus. With resolving power around 5 eV compared to ~100 eV for previous missions like XMM-Newton, subtle profile asymmetries and shifts become detectable. The question is whether we're interpreting them correctly.

Implications for Relativity Tests

Black hole systems serve as natural laboratories for testing general relativity. The metric around a rotating black hole—the Kerr solution—makes precise predictions about ISCO location, light bending, and frame-dragging. Spin measurements derived from iron line modeling are often cited as confirmations of these predictions.

But if systematic modeling biases have been underestimating spins, some high-spin measurements might be even more extreme than currently believed. This matters because spins approaching the maximum allowed value (a dimensionless spin parameter of 1) suggest sustained, coherent accretion—black holes being fed from a consistent direction over millions of years. Alternatively, it could indicate black hole mergers that happened to preserve high spin.

Systematic underestimation would also affect tests looking for deviations from general relativity. If the baseline model is wrong, apparent anomalies might simply reflect geometric effects rather than new physics.

The Corona Height Problem

The height of the X-ray corona above the disk affects light bending and illumination patterns. Current measurements typically place coronae at a few to tens of gravitational radii. But these measurements depend critically on disk geometry assumptions.

If thick disks cause us to underestimate corona height, it could resolve a puzzle: some systems show rapid X-ray variability that suggests compact coronae, while spectral modeling indicates large heights. Accounting for disk thickness might reconcile these tensions, suggesting coronae are actually more compact than spectral fits suggest.

Methodological Concerns

This is a preprint (dated April 2026, suggesting a future submission) and has not yet undergone peer review. The authors don't provide specifics about the range of parameters explored or statistical significance of the biases they identify—they simply state that biases occur "if fitted with a flat disk model" using "measurement uncertainties from XRISM."

Critically, we need to know: How large are these systematic errors compared to statistical uncertainties? For which types of systems (black hole mass, accretion rate, viewing angle) do the effects matter most? Can any observational signatures distinguish between different disk geometries?

The claim that warped disks "could not be fit" is particularly striking but underspecified. Does this mean reduced chi-squared values were unacceptable? That parameter estimates failed to converge? Or that the resulting parameters were physically unrealistic?

The Path Forward

Several developments could address these concerns:

Expanded spectral models: Incorporating realistic disk geometries into publicly available fitting tools like RELXILL or REFLIONX would allow observers to test whether their data prefer thick or warped configurations.

Multi-wavelength constraints: Optical/UV observations can independently constrain disk geometry through continuum modeling, providing priors for X-ray analysis.

Time-resolved spectroscopy: Variability patterns during flares or state transitions can illuminate how X-ray illumination varies as the corona height or geometry changes, breaking degeneracies.

Numerical simulations: General relativistic magnetohydrodynamic (GRMHD) simulations like those performed with codes such as HARM or Athena++ self-consistently calculate disk structure, magnetic field geometry, and coronal properties. Synthetic observations from these simulations can calibrate simplified analytical models.

Candidates for Re-analysis

Several well-studied systems might be prime candidates for re-analysis with improved disk models:

MCG-6-30-15: A Seyfert galaxy whose iron line has been analyzed extensively, showing complex profiles that have been interpreted as evidence for high spin and complex corona geometry.

1H 0707-495: Shows extremely broad iron line features and rapid variability, with spin estimates near maximal. Disk thickness could significantly affect these measurements.

Cygnus X-1: The first confirmed stellar-mass black hole, with recent spin measurements from NICER and NuSTAR. Thick disk models could test consistency across different observatories and methods.

Broader Context: The Art of Approximation

This research exemplifies a recurring challenge in astrophysics: balancing model complexity against computational tractability and data quality. Flat, thin disks are wrong—we've known this since Shakura-Sunyaev—but they've been "good enough" given observational limitations.

As instrumental capabilities advance, yesterday's acceptable approximations become today's systematic errors. We saw this progression with cosmic microwave background analysis, where increasingly precise measurements demanded more sophisticated foreground models and systematics control. We're seeing it now with gravitational wave astronomy, where third-generation detectors will require more accurate waveform models than current templates provide.

The black hole spectroscopy community now faces a similar transition. XRISM's resolution, combined with upcoming missions like Athena (if it survives budget pressures), demands that our models match our data quality.

What This Means for Einstein

General relativity has passed every test thrown at it for over a century, from Mercury's orbit to gravitational waves. Black hole spectroscopy represents one of the most extreme testing grounds—regions where spacetime curvature is maximal, where quantum effects might appear, where alternative theories of gravity make distinct predictions.

But these tests are only meaningful if we correctly account for astrophysical messiness. If we mismodel the disk, we can't cleanly separate relativistic signatures from geometric ones. The irony is that general relativity itself—through light bending and frame-dragging—creates the very geometric complexities that complicate its own verification.

Systematic biases from disk geometry don't necessarily threaten relativity, but they do threaten the precision of our tests. The difference between a spin parameter of 0.9 and 0.98 might seem small, but it could be the difference between ruling out or accommodating alternative theories.

The Bottom Line

This preprint argues that a foundational assumption in black hole astronomy—flat, thin accretion disks—may be systematically biasing our measurements of fundamental black hole properties. As X-ray observatories reach unprecedented precision, these geometric subtleties matter.

The implications extend beyond individual measurements to our understanding of black hole growth, jet physics, and tests of general relativity itself. If confirmed through peer review and validated against observations, this work suggests that decades of black hole spin measurements may need re-evaluation.

The good news: XRISM and future missions provide the data quality to detect and correct these biases. The challenge: developing, validating, and deploying more sophisticated models before systematic errors contaminate our expanding dataset of high-precision measurements.

In science, recognizing your assumptions is often the first step toward transcending them. For black hole astronomy, that reckoning may have just begun.