Enrique Garcia works on a supercomputer while wearing an oxygen tank on his back. The machine he looks after sits inside the technical building at ALMA’s Array Operations Site, on the Chajnantor plateau in northern Chile, 5,100 metres above sea level. It is the highest-altitude supercomputer in the world, housed in what is said to be the second highest building in the world, and the air around it is so thin that its cooling fans have to push twice the normal volume of air to keep the processors from overheating. The people who maintain it need help breathing too.
Nobody builds at that altitude for comfort. The Chajnantor plateau was chosen because it offered something rare: a large, very flat, very dry plain high in the Chilean Andes, where some areas record as little as 3 millimetres of rain in an average year and there is often almost no water vapour overhead. The price of that dryness is an altitude of about 5,000 metres, where oxygen masks often need to be worn and where there was little precedent for managing a large and technically complex construction project. The surrounding Atacama is the driest non-polar desert on Earth, averaging around 15 millimetres of rain a year, with some weather stations that have never recorded rain at all.
For astronomers working at millimetre and submillimetre wavelengths, water vapour is the enemy. It is the main gas determining how opaque the atmosphere is at those wavelengths, which is why so much effort goes into measuring the precipitable water vapour above a candidate site before anyone pours concrete. Systematic comparisons using identical instruments deployed on Maunakea in Hawaii, at the South Pole and around Cerro Chajnantor found conditions in the Chajnantor area clearly better than on Maunakea. Sitting above a large fraction of the atmosphere on the edge of the driest desert on the planet, ALMA looks through air that is frequently extremely low in water content.
What was eventually built there is probably the most ambitious ground-based astronomical project ever undertaken: a collaboration of 20 countries, extending over 16 kilometres at 5,000 metres elevation, a decade in construction and more than a billion dollars in cost. It has met or exceeded almost every scientific expectation set for it. One thing that did not match the predictions was the climate. The Atacama is normally one of the driest places in the world, but a site 5,000 metres up also gets the high winds and heavy snowfall that come with the altitude.
The altitude shapes everything about how the observatory is staffed. Iván López, ALMA’s safety manager, has described an operation of roughly 250 people, of whom about 50 are exposed to intermittent hypoxia, the medical condition in which the body does not receive enough oxygen. Workers travel from the Operations Support Facility at 2,900 metres up to the antennas at 5,100 metres in under an hour, and they make that trip daily, which is nothing like the slow acclimatisation that climbers use. The responses are practical and unglamorous: the technical building is now permanently oxygenated by an on-site liquid oxygen plant; backpacks were designed so that a worker can carry an oxygen tank anywhere; because the gas is very dry, staff are given nasal sprays to keep their airways moist; drivers on the road up and down are advised to take a co-pilot; staff work in teams of at least two, following exact procedures broken into small listed tasks; and everyone works a roster of eight days on site followed by six days off at sea level. The machinery suffers as well. Parts that were expected to last a certain number of years often last half that, and the time estimated for any given task has had to be revised. Mining companies face similar problems, but little of what they know has been published, so the observatory has learned as it went, drawing on staff who had worked in Hawaii and at the nearby APEX telescope.
In 2017 that thin, dry column of air became part of something larger. The Event Horizon Telescope links radio dishes across the planet using very long baseline interferometry, synchronising facilities with hydrogen maser atomic clocks and exploiting the Earth’s rotation to form, in effect, a single telescope the size of the planet, observing at a wavelength of 1.3 millimetres. The eight contributing telescopes stood on volcanoes in Hawaii and Mexico, mountains in Arizona and the Spanish Sierra Nevada, in the Chilean Atacama and in Antarctica. Each produced roughly 350 terabytes of data a day, stored on helium-filled hard drives that were then flown to specialised supercomputers called correlators at the Max Planck Institute for Radio Astronomy and at MIT Haystack Observatory to be combined. The resulting angular resolution, 20 micro-arcseconds, is enough to read a newspaper in New York from a café in Paris.
ALMA’s contribution was decisive. Adding it as an anchor station raised the sensitivity of the whole network by an order of magnitude, which is what turned the Event Horizon Telescope from an instrument that could detect structure into one that could make an image. “ALMA is the most sensitive facility in the EHT, and its 66 high-precision antennas were critical in making the EHT a success,” said Xavier Barcons, Director General of the European Southern Observatory.
On 10 April 2019, at coordinated press conferences around the world, the collaboration released the result in six papers published together in The Astrophysical Journal Letters: a ring of glowing plasma with a dark patch at its centre, the shadow of the supermassive black hole at the heart of Messier 87, a massive galaxy in the Virgo cluster. The object is 55 million light-years from Earth and has a mass 6.5 billion times that of the Sun. Its event horizon, the boundary from which the telescope takes its name, is about 2.5 times smaller than the shadow it casts and measures just under 40 billion kilometres across.
The image does not show the black hole itself. Black holes are black because no light escapes them, so the hole is invisible; what is visible is the hot gas swirling around it under the influence of gravity near the horizon. For decades these objects had lived in theory, in equations and in computer models of the sort used in films. What the array achieved was to move the event horizon out of mathematics and into the category of things that can be observed directly, and the plan is to keep going: new stations on other continents, and eventually satellites in space, for sharper images and possibly even moving pictures of the hot plasma.
For the people who had spent years predicting what such a ring should look like, the moment of comparison was uncomfortable. “The confrontation of theory with observations is always a dramatic moment for a theorist. It was a relief and a source of pride to realise that the observations matched our predictions so well,” said Luciano Rezzolla of Goethe Universität in Germany.