AAPPS Bulletin
> home > Feature Articles
First Results of the Event Horizon Telescope
K. Asada, M. Nakamura
File 1 : Vol30_No3_Feature Articles-6~11.pdf (0 byte)

DOI: 10.22661/AAPPSBL.2020.30.3.06

First Results of the Event Horizon Telescope



The Event Horizon Telescope (EHT) collaboration has revealed the first-ever images of a black hole shadow at the heart of a giant elliptical galaxy Messier 87 (M87). The EHT links ground-based radio telescopes around the globe to form an Earth-sized virtual telescope with an unprecedented highest angular resolution using very long baseline interferometry (VLBI) at millimeter wavelengths. Images visually reveal the strongest evidence of an existence of a black hole in the universe. The bright compact radio source with a diameter of 42짹3 micro-arcsecond (關as) suggests a supermassive black hole (SMBH) of (6.5짹0.7)횞 109 M(solar mass). An asymmetric ring-like morphology strongly suggests that we see gravitationally lensed emission from plasma rotating around the very vicinity of the SMBH event horizon. The image also supports the longstanding hypothesis that a SMBH powers an active galactic nucleus (AGN). The EHT collaboration demonstrates that VLBI at millimeter/sub-millimeter bands offers a powerful method to explorer gravity in its most extreme limit and at a previously inaccessible mass scale.


On April 10, 2019, the Event Horizon Telescope (EHT) collaboration released the "photograph" of a black hole into our visual world. This day becomes a memorable day in human history, because a black hole, first predicted by Albert Einstein over a century ago with his general theory of relativity, has been visually captured by the EHT collaboration. The EHT Collaboration, founded by Prof. Shep Doeleman at CfA/Harverd university, is an international collaboration established in order to image the shadow of a supermassive black hole (SMBH) for the first time. More than 200 researchers from 13 stakeholder institutes including the Academia Sinica Institute of Astronomy and Astrophysics (ASIAA) in Taiwan, the National Astronomical Observatory of Japan (NAOJ) and the East Asian Observatory (EAO) from the Asian region and more than 20 affiliated research institutes/universities participate this collaboration. While the EHT collaboration formally started in 2017, project planning and test observations were initiated in the early 2000s. In this report, we describe how the first image of a black hole was taken together, its impact and future prospects.

Black Hole and its shadow

Although we are unable to resolve the event horizon in the image, we expect to resolve a ring-like structure with a photon capture radius of Rc = rg, where rg GM / c2 is the gravitational radius of a black hole, in a nonrotating Schwarzschild [1] black hole. The photon capture radius is larger than the event horizon in the Schwarzschild metric, the so-called Schwarzschild radius RS ≡ 2 rg. Photons at radius < Rc are captured by the black hole [2], while photons at radius > Rc could escape to infinity. There is an unstable circular orbit at radius ≈ Rc that produces an enhanced bright emission due to gravitational lensing. For a Kerr [3] black hole with spin angular momentum, Rc varies with the dimensionless spin parameter a = J / Jmax, where JmaxGM2/c is the maximum value of the black hole angular momentum, and the cross section of the lensed photon ring departs from a true circle [4]. This change is very small for a low inclination angle (< 20째) with respect to the black hole spin vector up to ~2 % [5].

Event Horizon Telescope and its observations

Based on the theory of general relativity, the angular size of the shadow is simply determined by the mass (and spin) of the black hole together with its distance from the earth. The largest one is the BH of Sgr A*, which is at the center of our Milky Way Galaxy, and the second largest one is that of M87 in the Virgo Cluster. The apparent angular sizes are expected to be 50 and 40 횆sl211횆slmult0 (as), respectively. Therefore, we need to have a sufficient angular resolution in order to see the shadows of BHs. In keeping with this aim, Lo et al. [6] measured the size of the Sgr A* using Very Long Baseline Interferometry (VLBI) at 7 mm, and Shen et al. [7] pursued further finer structure by reducing the observing wavelength to be 3 mm. There is a growing recognition that further shorter millimeter/submillimeter (mm/submm) VLBI could be a unique technique to achieve a sufficient angular resolution to image the shadow of the black hole in the coming decade [8, 9]. The Greenland Telescope project is one of the innovations led by ASIAA, Taiwan, targeting the imaging of the shadow of the BH in M87 [10]. The project started at the end of 2000s, and has now joined the EHT collaboration [11].

VLBI is a technique that links radio telescopes across the globe to form a virtual telescope with an aperture that is equivalent to the size of the earth. Observing the same astronomical source simultaneously with all the mm/submm telescopes, signals from the object are recorded using specialized data recorders at each telescope site. Then, the data are sent to the correlator site, and weak astronomical signals are extracted by taking the correlation between the data taken at each site.


Fig. 1: EHT array used for 2017 observations together with Greenland Telescope, which started to participate in EHT observations from 2018.

From the early 2000s, experimental VLBI observations at 1.3 mm (230 GHz) using only a few telescopes in the mainland US and at Hawaii have been conducted and the results revealed that presence of the very compact structures those are equivalent with the sizes of the shadows of black holes in Sgr A* and M87 [8, 9]. Following these successes, the EHT collaboration conducted new observations aiming to make images of BH shadows in M87 and Sgr A* [12, 13]. M87 was observed on April 5, 6, 10 and 11, 2017 with 7 stations (ALMA, APEX, IRAM 30m, JCMT, LMT, SMA, SMT) located at 5 different sites on the Earth (note that the SPT cannot observe M87).

Data Calibration and Results

All the data taken at each site were sent to the MIT Haystack Observatory and the Max Plank Institute fur Radio Astronomy (MPIfR). Data were synchronized and correlated in order to extract astronomical signals recorded in the hard disks using a super computer. Information was reduced extensively in order to extract only essential information during this process. For instance, the total amount of data size was reduced to 1 TB from 5 PB. Once data were correlated, we calibrate the data for further analysis.

Three pipelines have been developed and used for the a prior calibration and extraction of the astronomical signals (fringe finding); the development and usage of one of the pipelines have been led by the East Asian data calibration team [14].

The first Black Hole Shadow Image

Fig. 2: The first image of a BH shadow taken by the EHT collaboration. (credit: EHTC, 12)

After a long process of careful data validation, calibration and many crosschecks by the EHTC, the first attempt to image the shadow of M87 was conducted in 2018 June/July. In order to avoid any possible biases, four independent imaging teams have been created, and they attempted to image the data without sharing any information among teams at the initial imaging stage. Two of the teams have been led by institutes in East Asia (the ASIAA in Taiwan and NAOJ in Japan). Three different imaging methods have been used. There were 4 days of the data for M87, and all the four teams imaged the black hole shadow image very similarly for all the four days of the data. Based on this similarity, it is concluded that the images we obtained were robust and had a very high reliability [15].


Fig. 3: Estimated size of the ring (EHTC, 17). Both image and visibility domain fitting for all the epochs using different methods give consistent values.

After that, the imaging teams collaboratively refined the image and the final 1.3 mm VLBI image of M 87 revealed asymmetric ring morphology of the central compact radio source with a diameter of 42짹3 關as [16]. The ring is almost circular and encompasses a central brightness depression of > 10:1. The robustness of this feature over multiple days strongly suggests that we see gravitationally lensed emission originating from near the black hole event horizon.

Interpretations of the Asymmetric Ring

The theory working group of the EHT collaboration built physical models in order to reproduce the observed images by utilizing massive numerical simulations [17]. This helps to understand the behavior of the plasma under the influence of the intense gravity in the vicinity of the black hole. General relativity suggests that the orbital paths of photons, which originated from the accreting gas, are eventually captured by the gravity of the black hole. Therefore, there is a dark region where the intensity is much lower than the surrounding area within a few gravitational radii (rg) including the black hole itself (inside the event horizon). Beyond this dark region (we call it a "shadow"), the so-called photon ring with an angular radius of rg (in reality, it is a photosphere) where photons will survive to rotate several times (gaining in intensity) and finally arrive at the earth. The EHT took photographs of the brightest photon ring around the SMBH with a depressed central darkness (the black hole shadow).

M87 is one of the typical sources that is categorized as a low-luminosity AGN and is also well-known to possess a jet. The jet from the nucleus of M87 travels more than 5,000 light years beyond the central region of the host galaxy. It is considered that a very hot (electron temperature of ~ 1011 K) accretion flow surrounds the SMBH in the center of M87 with dominant synchrotron radiation at millimeter/sub-millimeter wavelengths. The simulation library consists of general relativistic magnetohydrodynamic (GRMHD) simulations together with general relativistic radiative transfer (GRRT) calculations. They are essential tools for modeling the accretion flow and formation of jets around the black hole.

The theory working group performed extensive simulations using a wide variety of parameters including the black hole angular momentum, the event horizon-threading magnetic flux, and the electron temperature of the accretion flow. In total, more than 60,000 snapshots were used for cross checking with observed images during the EHT2017 campaign.

Overall the structure of simulated images reproduce the asymmetric photon ring with a diffusive component of the accretion flow and a funnel wall (outflow) which surrounds the funnel jet from a spinning black hole (Fig.4). Furthermore, it was beyond our expectation, but there are many models having different parameters that are consistent with the observations. This implies that the observed emission of the asymmetric ring does not depend much on the dynamics and/or structure of the accreting material, but it strongly indicates that the behavior of the photons can be described in the curved spacetime of GR. We note that a non-rotating black hole is ruled out because it will not support the jet power apparent in observations (≥ 1042 erg s-1). In this case (the black hole in M87 is spinning), the spin vector axis is uniquely determined as pointed away from Earth and the asymmetric emission (the lower part is brighter than upper part) is produced primarily by Doppler breaming: the bright region corresponds to the side approaching us.


Fig. 4: An example a of self-consistent GRMHD + GRRT simulated image (left). The model image convolved with a 20 關as FWHM Gaussian beam (right) (credit: EHTC, 16).

The theory working group adopted the working hypothesis that the central object in M87 is a black hole described by the Kerr metric. Under this assumption, observed images are pretty much consistent with our GR-based numerical models. However, it is interesting to consider whether or not observed images are also consistent with alternative models for the central object such as black hole "mimickers", i.e., compact objects, both within GR or in alternative theories. For example, regular horizonless objects without a surface are boson stars [18], while mimickers with a surface are gravastars [19]. A comparison of EHT2017 data with both the boson star model (as a representative horizonless and surfaceless black hole mimicker) and the gravastar model (as a representative of a horizonless black hole mimicker) would be important [20]. Further examinations will bring a strong constraint on the spacetime property around the central object in M87.

The asymmetric ring in EHT2017 also provides a robust estimation of the black hole mass by extracting characteristic properties, such as size and degree of asymmetry from the geometric model, GRMHD simulation model, visibility data, and reconstructed images. We find that > 50% of the total flux at arcsecond scales comes from near the event horizon with a dramatically suppressed interior emission by a factor > 10, providing solid evidence of a black hole shadow. Across all methods above, which are consistent with EHT2017 data, a diameter of 42짹3 關as was obtained. Folding in a distance measurement of 16.8 Mpc gives a black hole mass of (6.5짹0.7) 횞 109 M. This supports the value obtained by the stellar dynamical measurements [21].


Fig. 5: The structural profile of the M87 jet [34]. The structural transition was found by Asada & Nakamura [24] for the first time and observations were conducted during recent decades.

Future Prospects

Throughout the EHT2017 campaign running for ~ a week, the asymmetric ring structure was stably imaged, while unexpectedly we could not capture the innermost emission of the jet. How/where is the jet initiated? This long-standing question has remained unsolved for over a century since the astrophysical jet was found in M87 in 1918 [22]. It may be due to a lack of the sensitivity of current arrays (the expected photon flux from the jet and the accretion flow could be orders of magnitude lower than that of the photon ring). In 2018, the ASIAA-led Greenland telescope (GLT) joined the EHT observations. Furthermore, two more telescopes are planned to join after 2021. We therefore expect to image the launching regions of astrophysical jets for the first time.

During recent decades, East Asian (EA) astronomers have conducted various research programs in relation to M87 and play a major role in the community; Hada et al. [23] measured the VLBI core shift using multi-frequency VLBI observations and nailed down the possible location of the black hole in the upstream of the core at 43 GHz. Asada & Nakamura [24] explored the parabolic jet structure and discovered the jet collimation break at around the Bondi radius (see also, 25, 26, 27, 28). Kino et al. [29] argued that the innermost magnetic field at 230 GHz was associated with the jet base. Asada et al. [30] found possible spine jet emission in VSOP (VLBI Space Observatory Programme) observations (see also, 31 for a theoretical interpretation). Hada et al. [32] and Park et al. [33] studied the jet kinematics by utilizing the East Asia VLBI Network (EAVN). Nakamura et al. [34] investigated the parabolic jet from the spinning black hole using GRMHD simulations and applied it to the M87 jet. Takahashi et al. [35] analyzed a synthetic synchrotron map with a force-free jet model and constrained the BH angular frequency. Kim et al. [36] analyzed the nuclear spectrum with the Korean VLBI network and suggested the possibility of its flatness (indicating a strong magnetic field near the black hole). Park et al. [37] measured the Faraday rotation obtaining an indication of winds from hot accretion flows confining the parabolic jet.


Fig. 6: The velocity profile of the M87 jet [33]. A complicated profile was revealed by utilizing the EAVN facilities.

Fig. 5 represents one of extensive efforts led by EA astronomers (Nakamura et al. 2018). They compiled the jet emission at frequencies, ~ 1 GHz to ~ 230 GHz, to reveal the jet collimation structure in M87 and compared it with the theoretical model. Based on this effort, the EHT collaboration will seek an origin for the M87 jet with connecting to lower frequency emissions. The EAVN large program is currently conducting a very high cadence monitoring of the M87 jet (a few day to a week) and this is the one and only program that can investigate how/where the jet is accelerated to a relativistic speed. The latest result is summarized by Park et al. [33] as shown in Fig. 6. The velocity profile suggests a gradual acceleration of the jet and that it transits into a deceleration beyond the Bondi radius. This result is consistent with Fig. 5 and we speculate that the jet acceleration and collimation take place in a co-spatial manner, supporting the MHD jet paradigm. Thus, EA astronomers are very active in determining the acceleration and collimation in M87, which is one of the most fundamental quests in the field of astrophysical jets. Besides, EAVN observations are scheduled quasi-simultaneously with the EHT campaign. This provides the jet position angle and other important information in the downstream, giving a further constraint on the horizon-scale modeling of observed EHT images.

After 2017, the EHT collaboration continued to observe M87 and further investigations will be conducted including time variation of the photon ring (the stability of the asymmetric ring structure). Also, polarimetric observations will reveal the magnetic field morphology on the horizon scale. Thanks to the GLT (improving a north-south baseline), the resolution along the transverse direction of the jet has been improved. The EHT collaboration is planning to conduct 0.8 mm (345 GHz) observations as well as space VLBI (next generation EHT). As a result of imaging the black hole shadow at 345 GHz, Kawashima et al. [38] proposed a constrain on the BH spin in M87.

We remark that observations of our galactic center Sgr A*, another EHT target for imaging a BH shadow, are now under analysis by the EHT collaboration. The mass ratio between Sgr A* and M87 is about three orders of magnitude. If we could image BH shadows in both sources, it will provide a solid milestone in radio astronomy showing that VLBI is a powerful tool for studying the extremely curved spacetime around a black hole in a visible way. It also indicates that the general theory of relativity can be examined in regions of the strongest gravitational fields around black holes using electromagnetic waves. There is no doubt that a big leap will be brought to BH astronomy by the success of the EHT project. Studying a BH shadow may also play a complementary role with testing the general theory of relativity in stellar mass black holes observed in gravitational wave astronomy.

EA-led EHT arrays such as JCMT (EAO), SMA, and GLT (ASIAA/CfA) play an important role in providing critical baselines for calibrating data and maximizing spatial resolution. JCMT and SMA are key to accurately capturing the structure and time variation that affect an interpretation of the BH shadow. EA regional efforts (EAVN high frequency observations at 230 GHz: SPART in Nobeyama, Japan and SRAO in Seoul, Korea) have also begun. Further rapid progress in BH astronomy is surely promising in the coming decade.

Acknowledgements: This research was conducted by the EHT collaboration, which is an international collaboration for imaging the black hole shadow.

KA is supported by the Ministry of Science and Technology of Taiwan grant MOST108-2112-M-001-051.


[1] K. Schwarzchild, Sitzungsberichte der K철niglich Preu횩ischen Akademie der Wissenschaften (Berlin: Deutsche Akademie der Wissenschaften zu Berlin) (1916).
[2] D. Hilbert, Nachrichten von der K철niglichen Gesellschaft der Wissenschaften zu G철ttingen—Mathematisch-physikalische Klasse (Berlin: Weidmannsche Buchhandlung) (1917).
[3] R. P. Kerr, Phys. Rev. Let. 11, 237.
[4] J. M. Bardeen, in Black Holes, ed. C. DeWitt & B. S. DeWitt (New York: Gordon and Breach), 215.
[5] C.-K. Chan, D. Psaltis, F. 횜zel, ApJ, 77, 13 (2013).
[6] K. Y. Lo et al. Nature 362, 38 (1993).
[7] Z. Q. Shen et al. Nature, 438 62(2008).
[8] S. S. Doeleman et al. Nature, 455, 78 (2008).
[9] S.S. Doeleman et al. Science, 338, 355 (2012).
[10] M. Inoue, J.C. Algaba, K. Asada et al. 2014, RaSc. 49, 564.
[11] M.-T. Chen, P. Raffin, P.T.P Ho, et al. SPIE 2018, 10700E, OHC.
[12] EHT Collaboration, ApJL, 875, L1 (2019a).
[13] EHT Collaboration, ApJL, 875, L2 (2019b).
[14] EHT Collaboration, ApJL, 875, L3 (2019c).
[15] EHT Collaboration, ApJL, 875, L4 (2019d).
[16] EHT Collaboration, ApJL, 875, L5 (2019e).
[17] EHT Collaboration, ApJL, 875, L6 (2019f).
[18] D. J. Kaup, Phys. Rev. 172. 1331 (1968).
[19] P. O. Mazur and E. Mottola, PNAS, 101, 9545 (2004).
[20] H. Olivares, Z. Younsi, C. M. Fromm et al., Phys. Rev. D submitted (arXiv:1809.08682).
[21] K. Gebhardt et al. ApJ, 729, 119 (2011).
[22] H. D. Curtis, PLicO, 13, 9 (1918).
[23] K. Hada, A. Doi, M. Kino et al., Nature, 477, 185 (2011).
[24] K. Asada and M. Nakamura, ApJL, 745, L28 (2012).
[25] M. Nakamura and K. Asada, ApJ, 775, 118 (2013).
[26] K. Hada, M. Kino, A. Doi et al., ApJ, 775, 70 (2013).
[27] K. Hada et al. ApJ, 817, 131 (2016).
[28] J.-Y. Kim, T. P. Krichbaum, R.-S. Lu et al. A&A, 616, A188 (2018a).
[29] M. Kino, F. Takahara, K. Hada et al., ApJ, 803, 30 (2015).
[30] K. Asada, M. Nakamura, H.-Y. Pu, ApJ, 883, 56 (2016).
[31] T. Ogihara, K. Takahashi, K. Toma, ApJ, 877, 19 (2019).
[32] K. Hada, J. Park, M. Kino, PASJ, 69, 71 (2017).
[33] J. Park, K. Hada, M. Kino et al., ApJ, 887, 147 (2019).
[34] M. Nakamura, K. Asada, K. Hada et al., ApJ, 868, 146 (2018).
[35] K. Takahashi, K. Toma, M. Kino et al., ApJ, 868, 82 (2018).
[36] J.-Y. Kim, S.-S. Lee, J. A. Hodgson et al., A&A, 610, L5 (2018b).
[37] J. Park, K. Hada, M. Kino et al., ApJ, 871, 257 (2019).
[38] T. Kawashima, M. Kino, K. Akiyama, ApJ, 878, 27 (2019).



Keiichi Asada is an associate research fellow at the Academia Sinica, Institute of Astronomy & Astrophysics (ASIAA) in Taiwan. After receiving a D. Sci. from the Graduate University for Advanced Studies, he worked as a postdoc at the National Astronomical Observatory in Japan, Institute of Space and Astronautical Science (ISAS), the Japan Aerospace Exploration Agency (JAXA) and Academia Sinica, Institute of Astronomy and Astrophysics. After that, he was appointed as an assistant research fellow, then associate research fellow at the same institute. He is a project scientist of the Greenland Telescope Project and a member of the Event Horizon Telescope (EHT) Project. He also serves as a Science Council member for the EHT.


Masanori Nakamura is an associate professor at the National Institute of Technology, Hachinohe College in Japan (since April 2020) and a visiting scholar at Academia Sinica, Institute of Astronomy & Astrophysics (ASIAA) in Taiwan. After receiving a D.Sci. from the Tokyo University of Science, he worked at NASA Jet Propulsion Laboratory/California Institute of Technology, Los Alamos National Laboratory and the Johns Hopkins University/Space Telescope Science Institute before joining ASIAA in 2010. His research fields are theoretical and computational astrophysical fluid dynamics and high energy astrophysics. He is a member of the Event Horizon Telescope Project.

AAPPS Bulletin        ISSN: 2309-4710
Copyright 짤 2018 Association of Asia Pacific Physical Societies. All Rights Reserved.
Hogil Kim Memorial Building #501 POSTECH, 67 Cheongam-ro, Nam-gu, Pohang-si, Gyeongsangbuk-do, 37673, Korea
Tel: +82-54-279-8663Fax: +82-54-279-8679e-mail: aapps@apctp.org