[SciShow intro Plays]

Reid: Space travel has taught us a lot about the universe. Over the years, we’ve sent people to the Moon, built a handful of space stations and space telescopes, and sent spacecraft to explore all kinds of different worlds in our solar system -- not to mention the thousands of satellites orbiting Earth right now.

But there’s a side of spaceflight that we don’t often hear about, probably because it’s the sort of thing governments like to keep to themselves: spy satellites. There’s a lot of information that still isn’t public, especially when it comes to the exact capabilities of today’s military satellites. But every so often, governments will declassify some part of their military satellite program’s history, and we’ll get to find out a little more about what’s been going on behind the scenes.

During the 1960s and 70s, for example, the US spent a lot of time trying to make spy satellites practical. But it was hard, because the photos still had to be stored on film. So even if the satellites took great pictures, how was the government supposed to get ahold of them? The US government, at least, tried to solve this problem with early satellite systems like Corona, Gambit, and Hexagon.

From 1960 to 1972, the Corona program launched over 140 bus-sized, cigar-shaped satellites into orbit, each equipped with one, and later two, huge, wide lens cameras -- sometimes nearly 3 meters long. The tubes were also stuffed with thousands of meters of specially modified film. The satellites passed over enemy targets -- mostly the USSR -- several times a day, usually taking between a day and a week to use up all their film. Then came the hard part: getting the photos back. Spent film was held in one or more special capsules, or film buckets. And eventually, the satellite would drop that bucket, which would start plummeting to Earth in a controlled fall. Once it re-entered Earth’s atmosphere and hit a safe height of about 18 kilometers, the bucket’s heat shields would pop off and its parachutes would deploy.

Now here comes the best part: Because no one wants a pile of super secret spy photos crashing a random picnic, drops usually occurred off the coast of Hawaii, where a plane would be waiting to snag the package in mid-air with a sort of dangling claw. Special crews spent months practicing these complicated retrievals -- though they sometimes still missed, because it turns out that it’s hard to catch a falling bucket in a sling. The film they did manage to catch was then meticulously examined by dozens of microscope-wielding analysts, and even though the resolution was nowhere near as good as it is today, experts could still gather tactical information, compare new images with older ones, and learn more about the landscape.

We don’t have access to quite as much information when it comes to today’s intelligence satellites, but they probably collect similar kinds of information -- just a lot more of it. The technology behind digital imaging and wireless communication has improved a lot since the 60s and 70s, so getting high-resolution photos from today’s satellites is much simpler. I mean, you can get a pretty good satellite view of almost anywhere in the world just with Google Earth.

So, if commercial images have resolution that high, could a spy satellite figure out what book you’re reading in your backyard? Well, probably not, because we just haven’t been able to build satellites with good enough cameras. The Federation of American Scientists estimates that at a one-meter resolution -- meaning, each pixel covers one square meter of ground -- you’d be able to recognize a vehicle as a vehicle, and not a shed or a tree or whatever. A 50 centimeter resolution would give you a decent chance of identifying that car as a minivan, while at 25 centimeters -- the highest resolution allowed by the US government for commercial satellites -- you might be able to recognize its make and model. With a 10 centimeter resolution you’d probably be able to get an actual description of the van, or see that it had a roof rack, but you probably still couldn’t read its license plate or make out the Buffy bobblehead on the dashboard. Whether we actually have the technology to take photos with a resolution of 10 centimeters is a separate question.

There’s some speculation that military and intelligence satellites may be getting photos with a resolution of 15 centimeters or better, but we don’t know for sure, because that stuff is all classified. But we do know that one of the biggest challenges when it comes to analyzing satellite imagery is actually processing all that data. Back in the Corona days, there weren’t that many photos, so humans were able to carefully sift through them all. Now? There’s way too much data for humans to process everything by hand, and computer software isn’t that great at recognizing small objects in photos, or knowing when a change in a landscape might be important. But at least we aren’t swiping buckets out of the sky anymore.

Thanks for watching this episode of SciShow Space, and thanks especially to our patrons on Patreon who help make this show possible. If you want to help us keep making episodes like this, just go to patreon.com/scishow to learn more. And don’t forget to go to youtube.com/scishowspace and subscribe!