First Sagittarius A* Event Horizon Telescope Results. I. The Shadow of the Supermassive Black Hole in the Center of the Milky Way

To characterize the spectrum and variability of Sgr A∗, the EHT observing campaign was supported by parallel observations at several other observatories. Longer-wavelength VLBI observations, where interstellar scattering prevents direct observation of structural changes, were arranged to occur within a few days of the EHT campaign. IR and X-ray observations were arranged to be as simultaneous as possible with the EHT tracks on several days, and they resulted in one confirmed X-ray flare on April 11.

For the observations of April 6 and 7, when no strong flares were observed in the parallel observations, we must evaluate whether Sgr A∗ has structural variations within our observation periods. The simplest evidence for variability at λ=1.3mm comes from analysis of the total flux density (the “light curve”), which is measured during EHT observations using the ALMA and SMA connected element arrays. As discussed in Wielgus et al. [120], the fractional variability across all days is approximately 9%, with variations of 4−13% seen within individual nights (see, e.g., Figure 2). This variability is an order of magnitude stronger than is expected from interstellar scintillation. Thus, the light curves provide evidence for intrinsic image variability in Sgr A∗.

Each EHT baseline provides information about the detailed structure of this variability on an angular scale determined by the baseline length. On several triangles, EHT closure phases show slightly more variation than is expected from their thermal noise, the changing of projected baselines (from the source view) with Earth rotation, and interstellar scintillation. Comparison of interferometric visibility amplitudes for nearby baselines also reveals variability that significantly exceeds what is expected from thermal noise, calibration uncertainties, and baseline evolution.

To quantify the variability, we developed a simple parametric model for the spatio-temporal power spectrum for the variability of Sgr A∗. This model represents the variance in the visibility amplitude as a function of the radial distance in the (u,v)-plane, taking the form of a broken power law, as motivated by studies of GRMHD simulations. The source-integrated light curve is divided out in order to isolate structural variation from overall changes in flux density. The variability power spectrum was also empirically estimated by analyzing variations in visibility amplitudes on nearby baselines. This comparison included the same baselines sampled on different days, as well as nearby or crossing baseline tracks (e.g., SPT-LMT and SPT-SMA sample nearly identical baselines, but at different times). This analysis revealed that the fractional variability can be order unity for EHT baselines located near the two deep visibility minima (Figure 2), even after normalizing the data to remove light curve fluctuations, and that it significantly exceeds the variability expected from interstellar scintillation.