Black hole ring images from PSF structures

TL;DR

This paper investigates whether the ring-shaped images of black holes produced by the Event Horizon Telescope are genuine or artifacts caused by the shape of the point spread function, emphasizing the importance of PSF analysis in interferometric imaging.

Contribution

The study highlights the potential for PSF structures to create false ring images in VLBI data and emphasizes the need for thorough PSF deconvolution assessment in high-frequency interferometric observations.

Findings

Black hole ring sizes match PSF sidelobe intervals.

PSF artifacts can mimic black hole shadows.

Standard imaging techniques may not fully remove PSF effects.

Abstract

Two critical aspects of radio interferometric imaging analysis are data calibration and deconvolution of the point spread function (PSF) structure. Both of these are particularly important for high-frequency observations using a VLBI network consisting of a small number of stations, such as those conducted by the Event Horizon Telescope (EHT). The Event Horizon Telescope Collaboration (EHTC) has presented images of ring-shaped black holes from observations of M 87 (d = 42 +- 3 muas)(EHTC2019a) and the Galactic Center (d = 51.8 +- 2.3 muas)(EHTC2022a). The ring structures seen in the EHTC images are consistent with the estimated shadow diameter of the black hole based on its mass and distance. However, these black hole ring sizes are also the same with the typical up and down spacings (e.g., the intervals between the main beam and nearby 1st-sidelobes) seen in the point spread function (PSF; dirty beam) for each observation. These facts suggest that the EHTC ring structures are artifacts derived from the shape of the PSFs rather than the intrinsic structure of the SMBHs in M 87 and the Galactic Center. The EHTC utilizes novel imaging techniques in addition to the standard CLEAN algorithm. The CLEAN method was designed for PSF shape deconvolution in mind, yet in practice, it may not always be able to completely remove the PSF shape. In the imaging analysis of data from interferometers with a small number of antennas like the EHT, it is crucial to assess the PSF shape and compare it with the imaging results. The novel imaging methods employed by the EHTC have not yet been fully evaluated for PSF deconvolution performance, and it is highly recommended that their performance in this regard be thoroughly examined. It is also important to investigate the data calibration capability, i.e., the ability to separate error noise from the observed data.