Dec 30, 2018 1:00 AM

How oil traders make big bucks by using satellite surveillance

A flurry of Cubesat launches is already changing economic forecasting. In the near future, it could have an even more profound impact

Chris Hadfield wowed us with a zero-gravity performance of “Space Oddity”; Elon Musk showed us his enormous rocket; but in 2019, the new poster child for the space industry may be something rather more understated – a satellite the size of a Beanie Baby box.

Satellites used to be huge: many tonnes, several metres tall, and with solar panels the size of a bus. But 20 years ago, two engineering professors, Robert Twiggs and Jordi Puig-Suari, decided that their graduate students needed a constraint: they challenged them to build a satellite no bigger than, well, than a Beanie Baby box. (Professor Twiggs had seen one on a visit to the shops.)

That challenge has set the standard for nanosatellites: a CubeSat unit, 10cm by 10cm by 11.35cm. Many CubeSats are two or three units long – about the size of a toaster, and packed with off-the-shelf components from smartphones. And in 2019, these CubeSats are poised to revolutionise economics.

CubeSats are cheap to make – and just as important, cheap to launch. Building and launching a major satellite would have cost half a billion dollars not long ago. You could get a CubeSat into low Earth orbit for closer to $100,000.

One Silicon Valley company, Planet Labs, was founded in 2010, but already has the world’s largest private fleet of satellites – more than 200 of them, mostly CubeSats. By 2019 they are likely to be taking a million high-resolution photographs a day, covering anywhere on the globe that anyone might want to photograph, every 24 hours.

At the front of the queue for these photographs are the commodity traders. The NPR economics podcast Planet Money – which evidently could not resist the opportunity to team up with a company called Planet Labs – found a striking example. Armed only with high-resolution photographs from space, oil traders could measure how full floating oil tanks were in ports across the world, by measuring the length of their shadows and adding a spot of trigonometry. This modest edge in understanding the flow of oil can be worth serious money. The same can be said for wheat, coffee, soya – anything. This will take economic forecasting to a new level, and the technology will be mature by 2019.

My inner nerd, however, hopes for something a little bit more than an edge in the commodity markets. Almost 25 years ago, an economist named Alex Pfaff realised that satellite images could answer questions about the connection between development projects and deforestation in the Amazon. It was probably the first economic question to be answered by images from space.

Other economists have used night-light images to measure economic growth. They track air pollution to answer questions about the impact of road pricing or even the construction of metro stations.

Now, machine-learning means that the sudden surge in the volume of images from private CubeSat fleets is manageable. Algorithms can ask questions such as how many of those Brazilian favela dwellings have new roofs? What percentage of roads in Uganda are in good shape? Economists are now frantically scrambling to sharpen their skills in handling these tools.

Science has always depended on tools that help us measure and perceive – from Galileo’s telescope to LIGO, the laser interferometer gravitational-wave observatory. Now economists have their hands on some new tools that even a physicist might admire. There’s so much going on under the surface of a big economy, and so much that doesn’t show up in regular statistical releases for months – sometimes years. In 2019, we’ll be able to see it day by day.

Tim Harford is the “Undercover Economist” at the Financial Times, and the author of Fifty Things That Made the Modern Economy