The high-res pictures on Google Earth are taken from aircraft, not from space.

“Hubble…is capable of imaging with the _shortest_ exposure time of 0.1 sec. For Hubble in a case where it’s attempting to image something on Earth, anything in the FOV would move ~700 meters before the image shutter closes. The satellite and any like systems are simply not capable of establishing the requisite pointing lock.”

Typical Hubble targets are many orders of magnitude fainter than the surface of the earth, so a faster shutter on Hubble wouldn’t make a lot of sense. Large diameter optics on a spy satellite increase resolution, but they also shorten exposure times (so orbital motion is less of a problem). The fastest high-speed cameras are capable of capturing images with durations as short as 4e-8 sec. At that speed, image blurring would not be a problem.

My argument was entirely about what’s possible given the _optics_ of a Hubble-sized KH-11. I wasn’t claiming that KH-11s were _mechanically_ identical to Hubble (in fact, Hubble can’t focus on Earth–its near focus is greater than its orbital altitude).

“Compare this to the work to collect the Hubble Ultra Deep Field.YD3 image: HST held a precise, accurate lock on its target and kept the shutter open for something like 11.6 consecutive days (yes, days).”

I’m well aware–I was on the team responsible for planning and scheduling the first HDF (which, BTW, was not a multi-day long, single exposure–it was many shorter exposures stacked together. The trickiest thing to maintain was HST’s roll orientation throughout, not its pointing.)

But comparing (relatively) long exposures of dim, deep space targets to short snapshots of bright terrestrial ones is apples and oranges. The game with a spy sat isn’t to maintain lock over long periods (which wouldn’t help even if you could because the of the constantly changing angle at which the target is being viewed), but to take very short snapshots to “freeze” the motion, much like a sports photographer. I guarantee a KH-11′s fastest “shutter speed” is a lot shorter than 0.1 sec.

What would one of these people have paid for this picture? How much could they have learned from just this one image? Looking down, I see the way the clouds follow the land differently than the ocean, they way they form striations across distances, etc. etc. I can see things they could only imagine.

The world is full of amazing things, you just need to remember your perspective.

= = =

Large circular-ish undepression… that’s the Adirondacks.

Anyway, go watch that video of the ISS flying over the earth at night and then tell me it’s humbling or mind-blowing to see how undetectable we are.

I appreciate your reasoning but Earth-imaging resolution is not as straightforward as you estimate. A system like HST can’t collect surface images primarily b/c of Earth’s rotational velocity. For this reason it’s not a good “design reference”.

Hubble flies somewhere in the neighborhood of 550km nominally. It is capable of imaging with the _shortest_ exposure time of 0.1 sec. This is actually a really long time when talking about cameras. For Hubble in a case where it’s attempting to image something on Earth, anything in the FOV would move ~700 meters before the image shutter closes. The satellite and any like systems are simply not capable of establishing the requisite pointing lock.

Compare this to the work to collect the Hubble Ultra Deep Field.YD3 image: HST held a precise, accurate lock on its target and kept the shutter open for something like 11.6 consecutive days (yes, days).

@MadScientist – A thought experiment: One can have an effective resolution of 40030 km (size of Earth’s disk) and still find signs of humans. Just collect a series of night side images of the planet. City lights would change the temperature of the lone significant pixel.

It’s actually fairly straightforward to estimate:

The largest surveillance satellites (as far as we know) were designed to be taken into space by the shuttle (later ones have been launched by Delta IVs). The shuttle’s payload bay limits the aperture to something similar to Hubble (2.4 m). (In fact, from what I understand, a good portion of the reason Lockheed-Martin got the contract to build Hubble’s main architecture was because they already had experience in building similar-sized, space-based telescopes, AKA the KH-11 or “keyhole” satellites.) These birds generally have a perigee of around 250-300 km. The theoretical resolution of a 2.4 m mirror for (say) yellow light (580 nm) is about 0.05 arcsec.

At a distance of 250 km, that corresponds to a resolution of about 6 cm. That’s good enough to distinguish a man from a woman, a bald man from one with hair, or a man with facial hair from one who is clean shaven (in other words “Patriot Games” pretty much got it right). Probably not good enough to read a license plate, although with multiple images from multiple angles, you might be able to tease out that scale of detail with some fancy image deconvolution algorithms.

Now, since unmanned heavy launch vehicles have taken over for the shuttle, the 2.4 m mirror is no longer a limiting factor, so possibly there are bigger birds up there (although probably not significantly bigger). And, by using shorter wavelengths (near-UV?) that can still get through atmosphere, you should be able to tighten things up further. So yeah, it’s a pretty safe bet that if the military wants to read your license plate from orbit, they can. But they’re probably nowhere near reading the VIN off the dashboard plate, or being able to distinguish you from someone with roughly similar facial features–yet.

= = =

Farmland? Contrails? And the resolution isn’t actually all that hot.

Yes. But if they told you, they’d have to kill you.

Congratulations to the National Polar-orbiting Operational Environmental Satellite System Preparatory Project (NPP) team, all those who designed, built and fly her. Great job, superb work, thankyou.

But any new “eyes” on Earth are good to have, and this missions sensors are tailored for its specific mission, for which the above-menitoned sensors are not as well suited.

CJSF