Thread by @SaraIssaoun

A year ago today, we unveiled the M87 #BlackHole image by @ehtelescope! We were so proud to finally share our hard work (by 300+ scientists & engineers) with the rest of the world! Let me tell you a little bit about the crazy journey to get here
1/n
#OTD #BlackHole1year 🥳

@ehtelescope uses the technique of very long baseline interferometry to achieve the resolution necessary to see a black hole 55 million light-years away! We synchronized 📡📡 around the world, and synthesized a virtual telescope the size of Earth!
cr. @TheNRAO

The possibility to see the shadow of a black hole with VLBI was first put forward by my advisor, @hfalcke, back in 2000. At just the right radio wavelength, the gas flow around the black hole would be thin enough to peer down into the darkness!

cr. Falcke, Melia, Agol 2000

How big must our telescope be to see a black hole's shadow? The smaller an object appears on the sky, the higher the resolution we need. Since resolution is proportional to telescope size, higher resolution means bigger telescope!

There are two black holes close and massive enough for us to see from Earth: one at the center of the M87 galaxy, and one at the center of our own Milky Way. To be able to resolve them, we need a resolution of about 20 micro-arcseconds, meaning we need an Earth-sized telescope!

A decade long effort, led by our founding director @ShepDoeleman was then started to construct a 📡 network to observe at that wavelength (1.3 mm). To achieve the strongest data quality, we needed telescopes at high and dry sites: atmosphere and water vapor destroy our signal.

Members of @ehtelescope equipped observatories with receivers and recording equipment like this one to achieve the highest data quality. Each telescope records its data onto stacked hard drives, which are then combined at a later time.

cr. P. Torne

In 2017, we finally had enough telescopes to maybe obtain an image: the more we have, the more filled our 'virtual mirror' becomes. Our observer teams then shipped out to telescopes in April for our campaign (I was in Arizona!) We carried out 5 days of observations, 4 of M87!

The recorded data then got shipped to our 'correlators' at @MITHaystack in Massachusetts and MPIfR @maxplanckpress in Germany: supercomputer facilities that combine all the data and realign the signals that arrived at each telescope, creating a virtual telescope the size of Earth

After correlation, the data then moved on to a small team of data calibration experts. We developed tools to further refine and validate data quality, and a long feedback process led to final reduced data ready for analysis!

cr. L. Blackburn

4 imaging teams used various software to reconstruct images. The teams were not allowed to talk to each other for a period of 7 weeks, where everyone had the opportunity to make their first black hole images. I co-led Team 2 with @sparse_k , shown here!

cr. S. Ikeda

After 7 weeks, we met at a workshop and were thrilled to discover that all teams had similar images: a ring of the same size, brighter on the bottom! This is the whole imaging group celebrating the success, the average of the 4 team images shown on screen!

To decide on the best image, we tested our algorithms on data where we knew the truth. We then used the methods giving the best results on M87. 3 softwares, eht-imaging by @thisgreyspirit , SMILI by @sparse_k , and classic Difmap were used. The famous 📸 is the avg of the 3!

Did you know we have more than one image of the M87 #EHTblackhole? @ehtelescope observed it for 4 days in 2017, we have four independent images of it! They all show a ring of the same size, brighter on the bottom! #BlackHole1year

Our dedicated team of theorists compared the data to hundreds of simulations of hot gas and magnetic fields in the strong gravity of a supermassive black hole. Here's our beautiful Simulation Library for M87!

cr. A. Broderick/PI, U. Illinois, G.U.Frankfurt, Radboud U., Harvard

Our team of modelers fitted geometric models and simulations to the data, and measured the size of the black hole shadow to be 42 micro-arcseconds! The size of a BH is proportional to its mass, so we measured the mass of M87's BH to be a whopping 6.5 BILLION solar masses!

Here's my favorite @xkcdComic to give you a sense of scale, the M87 black hole shadow is bigger than our entire Solar System! If you want to know more, check out my other threads where I tackle some questions we often get from the public about our result!

cr. Randall Munroe

Thanks for celebrating #BlackHole1year with us! 🥳
The true fruits of our labor are the 6 publications that came with the image, they contain a lot more information! This was just the beginning for @ehtelescope, there is more to come!

cr. @SaraIssaoun @froeloefs @mmosc_m