Observing—and Imaging—Active Galactic Nuclei with the Event Horizon Telescope

Vincent L. Fish, Kazunori Akiyama, Katherine L. Bouman, Andrew Chael, Michael D. Johnson, Sheperd S. Doeleman, Lindy Blackburn, J. F. C. Wardle, William T. FreemanGalaxies 2016 · DOI

TL;DR

The Event Horizon Telescope (EHT) is pushing the boundaries of imaging supermassive black holes with unprecedented angular resolution. This paper reviews the EHT's capabilities and highlights the development of novel image reconstruction algorithms, crucial for extracting detailed information from its sparse data, promising deeper insights into black hole physics and AGN jets.

What Problem Does This Paper Address?

Imaging the immediate vicinity of supermassive black holes, like Sagittarius A (Sgr A) and M87, presents a monumental challenge. These objects are incredibly distant, and their event horizons, while theoretically predicted to cast a shadow, are minuscule on the sky. To resolve these features, astronomers need extremely high angular resolution, far beyond what single telescopes can achieve. The Event Horizon Telescope (EHT) tackles this by creating a virtual telescope the size of the Earth using Very Long Baseline Interferometry (VLBI). However, VLBI arrays, by their nature, have sparse coverage of the "uv plane" (the Fourier domain of the sky). This sparsity means that standard image reconstruction techniques, like CLEAN, struggle to produce accurate and high-fidelity images, often leading to artifacts and limiting the resolution that can be reliably extracted. The paper addresses the need for advanced imaging algorithms that can effectively handle this sparse data and extract the most information possible, especially for polarimetric and high-resolution imaging.

Methods

The paper discusses several advanced imaging techniques developed or adapted for the EHT, moving beyond traditional CLEAN:

  • CLEAN: The standard algorithm for radio interferometry, which iteratively subtracts the "dirty beam" from a "dirty map." While effective at lower frequencies, it struggles with the EHT's sparse uv coverage and large sidelobes.
  • Polarimetric MEM (Maximum Entropy Method): This method leverages maximum entropy principles, favoring sparse solutions, and has been extended to handle robust VLBI observables like closure phase and polarimetric ratios. It's shown promise for reconstructing polarized structures in AGN.
  • Bi-Spectrum Sparse Modeling (BSSPM): This technique uses compressed sensing, specifically the LASSO regularizer, to achieve superresolution. It's adapted to work with closure phases and incorporates a Total Variation (TV) regularizer to preserve sharp edges, making it suitable for imaging features like black hole shadows and jets.
  • Continuous High-resolution Image Reconstruction using Patch priors (CHIRP): This method breaks images into patches and uses data-driven priors learned from training on various image datasets. It aims to encode natural image statistics into the reconstruction process, potentially leading to more robust and higher-fidelity images, especially in low signal-to-noise scenarios.

Datasets

The paper discusses the application of these imaging techniques to data from the Event Horizon Telescope, which observes at 1.3 mm and is being extended to 0.87 mm. Specific targets mentioned include:

  • Sagittarius A (Sgr A): The supermassive black hole at the center of the Milky Way.
  • M87: The supermassive black hole in the giant elliptical galaxy Messier 87.
  • Quasars 3C 279 and 1924-292: Used for testing and validating imaging techniques.

The paper also references validation on simulated EHT data and real data from other VLBI arrays like the Very Long Baseline Array (VLBA) and the Global mm-VLBI Array (GMVA). The development of a beamformer for the Atacama Large Millimeter/submillimeter Array (ALMA) is highlighted as a crucial step in enhancing the EHT's sensitivity and baseline coverage, enabling observations of a wider range of sources.

Key Insights and Ideas

  • The EHT's Sparse Data is a Major Hurdle: The limited number and distribution of telescopes in the EHT array result in very sparse uv coverage, making standard imaging algorithms like CLEAN insufficient for high-fidelity reconstruction.
  • Robust Observables are Key: Due to atmospheric effects, the EHT relies on "robust" VLBI observables like closure phase and polarimetric ratios, which are less sensitive to instrumental and atmospheric variations. New algorithms must be able to effectively utilize these.
  • Beyond Deconvolution: Forward Modeling and Data-Driven Priors: Newer techniques are moving towards forward-modeling approaches (evaluating images in the data domain) and incorporating sophisticated, data-driven priors (like those learned by CHIRP) to guide the reconstruction process, rather than solely relying on deconvolution.
  • Superresolution is Achievable: Algorithms like BSSPM show promise in achieving angular resolution finer than the fringe spacing of the longest baselines, crucial for resolving the fine details of black hole shadows.
  • Polarization Imaging is Crucial: Understanding the magnetic field structure around black holes requires accurate polarimetric imaging, and techniques like Polarimetric MEM are being developed to meet this need.

Key Results

  • The EHT has successfully detected and provided size estimates for the emission regions around Sgr A* and M87, finding them to be only a few gravitational radii in size.
  • These observations, combined with weak infrared emission, strongly support the existence of event horizons for these sources.
  • New imaging algorithms like Polarimetric MEM and BSSPM have demonstrated superior performance compared to CLEAN on simulated EHT data, achieving higher fidelity and effective angular resolution.
  • CHIRP has shown promise in outperforming other algorithms, particularly for lower signal-to-noise datasets, by leveraging learned image priors.
  • The inclusion of ALMA is expected to significantly enhance the EHT's sensitivity and baseline coverage, enabling better images and extending the range of observable sources.

Limitations and Future Work

  • Limitations of CLEAN: The paper explicitly states that CLEAN is less well-suited for EHT data due to sparse uv coverage, large sidelobes, and the need for robust VLBI observables.
  • Prior Assumptions: While newer methods incorporate priors, the paper acknowledges that the "ideal" priors for black hole images are not yet fully understood, and arbitrary choices could introduce biases.
  • Future Work: The paper emphasizes the ongoing development and validation of new imaging algorithms. Future work will involve applying these techniques to more extensive EHT datasets, including polarimetric data, to further refine our understanding of black hole environments and AGN jets. The integration of ALMA is a significant future step that will expand the EHT's capabilities.

Summary updated

Abstract

Originally developed to image the shadow region of the central black hole in Sagittarius A* and in the nearby galaxy M87, the Event Horizon Telescope (EHT) provides deep, very high angular resolution data on other active galactic nucleus (AGN) sources too. The challenges of working with EHT data have spurred the development of new image reconstruction algorithms. This work briefly reviews the status of the EHT and its utility for observing AGN sources, with emphasis on novel imaging techniques that offer the promise of better reconstructions at 1.3 mm and other wavelengths.
BibTeX
@inproceedings{fish2016observingand,
  author = {Vincent L. Fish and Kazunori Akiyama and Katherine L. Bouman and Andrew Chael and Michael D. Johnson and Sheperd S. Doeleman and Lindy Blackburn and J. F. C. Wardle and William T. Freeman},
  title = {{Observing—and Imaging—Active Galactic Nuclei with the Event Horizon Telescope}},
  booktitle = {Galaxies},
  year = {2016},
  doi = {10.3390/galaxies4040054},
}

Loading related content...

Figure