Filming a black hole movie faces a fundamental problem: light from different parts of the accretion disk takes different times to reach the observer, and if this is ignored, the image becomes distorted. Existing fast approximations read the same plasma snapshot for all rays, which works for compact sources but fails for extended jets and polarization. The full slow calculation requires holding hundreds of three-dimensional snapshots in memory simultaneously, which exceeds the capabilities of ordinary computers. This work joins efforts on hybrid approaches, where the slow calculation is applied only to part of the space, but until now there was no way to know in advance which part to choose.
The authors created a software framework called CoportSL that first performs a cheap preliminary analysis: for each region of space, it estimates the contribution to emission, absorption, and Faraday effects. Then it automatically identifies regions where the total contribution exceeds a given threshold, and only there does it retain time dependence. For the rest of space, time is frozen, but radiative transfer is still computed along the entire ray. Additionally, the method constructs an effective time window from the distribution of propagation delays within the selected region, discarding rare tails. All this was tested on data from a magnetically arrested disk similar to M87*, with two electron models and two image scales: one close to the horizon and one wide including the jet.
For the near-horizon region, the method reduced the number of simultaneously loaded layers from 903 to 223, a 75.3% reduction, and for the jet from 897 to 496, a 44.7% reduction. The maximum error in Stokes images did not exceed 4×10⁻³ compared to the full calculation. Memory required for the main data structures dropped from 65.7 to 20.2 gigabytes for the near-horizon image and from 64.1 to 37.3 gigabytes for the jet, fitting within a workstation. The computation time per frame with the hybrid slow method remained comparable to the fast method at low core counts and was only 1.9 times longer on 64 cores. Comparison of fast and slow light curves showed high correlation for the near-horizon region, but for the jet, a global time shift does not eliminate local discrepancies, and spatial differences in polarization remain significant.
This work differs from neighboring hybrid schemes in that it is the first to determine the slow region not geometrically or by brute force, but by the contribution of the transfer coefficients themselves, including Faraday terms that affect polarization even where there is no bright emission. This directly answers the question of where fluid evolution must be retained for full Stokes. The method also, for the first time, links the choice of time window to the distribution of delays within the already selected region, rather than the full range. This combination allows quantitative error estimation and resource savings without repeated runs.
Further development could proceed by adapting the region partitioning and time window to a specific observed quantity, such as photon ring statistics or short-period variability. It will also be necessary to investigate the cause of parallel scaling saturation at high core counts and optimize data access for long sequences. Testing on other accretion states, viewing angles, and electron models will show how universal the found thresholds and windows are.
The method will directly affect the preparation of theoretical image libraries for the next-generation Event Horizon Telescope and the interpretation of dynamic polarimetry. It may also be useful for modeling blazar variability and other relativistic jets where the finite speed of light is significant.
The immediate next step is to convert fast and slow images into visibilities and closure quantities with realistic baseline coverage and noise, to understand which structural differences ngEHT can detect. Then, tests should be repeated for the photon ring and echoes, where relative delays themselves are the target.
In black hole astrophysics, an open question remains: how to distinguish light propagation effects from the intrinsic dynamics of the accretion flow. This work quantitatively shows for the first time that fast light can yield nearly identical light curves but strongly different spatial polarization maps, meaning that neglecting the finite speed of light can lead to incorrect interpretation of the magnetic structure near the horizon.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The work presents CoportSL, a method for automatically selecting the regions where plasma evolution matters for polarized radiation, which makes it possible to cut down on the number of snapshots loaded at once when modeling black hole images. The method relies on a preliminary analysis of the contributions from transfer coefficients and Faraday effects, and it also constructs an effective time window from the delay distribution. On data resembling M87*, it achieved a reduction of layers by 75% for the near region and 45% for the jet, with an error below 0.4%.
Where its strength is
The strength lies in the fact that the method is the first to define the slow region not geometrically but by the contribution of the transfer coefficients themselves, including the Faraday terms, which directly answers the question of where evolution must be retained for accurate polarization. This makes detailed modeling of dynamic images feasible on ordinary workstations.
Where it could go next
Worth trying
Why this may matter
The work opens the way to routine modeling of dynamic polarized images of black holes on workstations, which matters for interpreting future observations from the next-generation Event Horizon Telescope and for testing theories of accretion and relativistic jets.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.52, against a usual 0.64. It looks like few have drilled nearby.
For M87*, one gravitational radius is about 9.6 billion kilometers, and the time for light to cross it is roughly 8.9 hours. In the paper, delays are measured in these units, and even a 'short' window of 110 gravitational radii means a spread in emission times of almost 41 days.
In the movie 'Interstellar', time delay near the black hole Gargantua is shown as gravitational time dilation, but here it is different: light from different parts of the disk reaches the camera at different times, and without accounting for this, the image would not just be shifted but assembled from pieces of different ages, like a collage.