Disaster-proof astronomy?

The Atacama Large Millimeter/submillimeter Array (ALMA) on the Chajnantor Plateau in Chile. (Courtesy: Clem & Adri Bacri-Normier/ESO)

By Louise Mayor in San Pedro de Atacama, Chile

In many ways, the Chajnantor Plateau in the Chilean Andes seems like one of the worst places in the world to build a very large and expensive telescope array. I have already experienced or witnessed first-hand a host of hazards on my trip to the Atacama Large Millimeter/submillimeter Array (ALMA), which is my reward for winning the European Astronomy Journalism Prize 2014.

At 2.39 a.m. local time last Monday, I was rudely reminded that I was in a tectonically active region by a magnitude-6.3 earthquake. At the time, I was staying overnight in Santiago, with two flights down and one to go on my way to the ALMA site in the Atacama Desert further north.

It was quite a gentle awakening. The bed jiggled from side to side to a soundtrack of the rhythmic squeaking of the room’s lamp shades. I came to and remembered where I was – far from home in a hotel room in Santiago, Chile, and not too far from where there was a magnitude-8.3 earthquake a few days before my arrival. But by the time I had wondered if I should be taking some sort of safety measures, the shaking had subsided. I had got out of bed by this point and when I lay back down, I thought the shaking had started again. But I soon realized that the cause of this second round of shaking was my incredibly strong adrenaline-fuelled pulse.

So what magnitude of earthquake would it take to upset ALMA? I asked about this once I finally arrived and was in the control room of the ALMA Operations Support Facility, the site at an altitude of 2900 m where most of the staff work, nestled in the Andean mountain chain below the Array Operations Site at 5000 m, where the 66 antennae are sited. No-one I spoke to remembered the exact figure, but it was generally agreed that the telescope can withstand at least a local magnitude-8.0 earthquake, if not 8.3–8.5; the antennae were designed to withstand this amount of shaking. In addition, I was told, many of the antennae are on “floating” bases, which means they sit on soil rather than on rock, and so they are less vulnerable to ground vibrations.

Okay, I thought, that’s that eventuality taken care of. But whoa, what am I seeing on this screen here?! In the control room is a live video stream from an active volcano called Lascar, which is situated not too far from ALMA. Once I knew to look for it, I could see it with my own eyes, smoking in the distance. Lascar is apparently a big risk. Its last eruption was in 2006, and was big enough that the fire could be seen in the caldera along with an ash cloud. Another eruption would, of course, mean the evacuation of all employees. The danger to the telescope would be that the surfaces of the mirrors would be coated with corrosive and abrasive ash.

Another hazard, which hadn’t occurred to me until I spoke with electronics engineer Alejandro Sáez Medaim, is that at high altitude the electronics are more likely to be damaged by heavy particles from space, which can modify the content stored in the memory, or in the worst case cause a short-circuit. These “single event upsets” can happen daily and in most cases are harmless, but about once a month, one affects the correlator – the room full of electronics at the Array Operations Site where all the observation data feed in, referred to as the “brain” of ALMA. These monthly occurrences need a bit more work to fix, but this can be done remotely from the Operations Support Facility by reloading the affected device.

The thin air at high altitude also means that cooling fans are less effective at removing heat from the correlator electronics – a factor that is corrected for by passing an increased volume of that thin air through the correlator room.

So, what with all these hazards to worry about, why was the decision made to build ALMA on this earthquake- and volcanic-eruption-prone site at high altitude? Well, the key “pro”, and perhaps the most well known, is that at 5000 m above sea level, the telescope is above most of the water in the atmosphere. The bad thing about overhead water is that is absorbs the millimetre and sub-millimetre radio waves from space that astronomers are trying to detect using ALMA. In the control room, astronomers Violette Impellizzeri and Linda Watson showed me a screen indicating a “precipitable water content” of 1.5 mm. This means that if you compressed all the water above, it would have a depth of 1.5 mm. This did not surprise me, as I could feel that the air was very dry – the humidity was 2% up at the Array Operations Site.

A perhaps less-obvious advantage of the site is that the mountain plateau is so big that antennae can be placed 16 km apart. Such a long baseline can yield an incredibly high image resolution. The Altiplano – the high plain – found in this region is second in size only to the Tibetan Plateau.

What’s more, according to my tour guide, ESO Press Officer for Chile Francisco Rodriguez, big observatories have been built in worldwide consortia in Chile since the 1960s. That means that, in Chile, the industry of building major telescopes is well established – with others including the Very Large Telescope in Paranal, La Silla Observatory, the Cerro Tololo Inter-America Observatory and the Giant Magellan Telescope. This track record brings with it a whole load of experience and skills, and in fact 80% of the staff at ALMA are Chilean, my guide explains.

So perhaps, on reflection, the Chajnantor Plateau in the Chilean Andes is not one of the worst places in the world for a very large and expensive telescope array like ALMA, but it is in fact the very best.