6

A report broadcast on the TV BBC News Channel on 2018-11-01 at approximately 20:55 GMT (not available online) claims that soon, satellites will provide "daily 30 centimetre resolution global coverage". The report was made in an item on counting whales. I don't doubt that we can count whales from space if we're clever or lucky enough to point our best satellite instruments at the right spot at the right time, but global daily coverage at such a resolution would need a huge number of satellites and generate a truly huge amount of data — it doesn't sound credible. I believe they were referring to the Worldview satellite series. The WMO OSCAR database reports that the latest Spaceview-110 instrument has indeed a 31 cm resolution, but a 13 km swath width and global coverage in about six months, which means that daily global coverage would need about 180 satellites. That is a lot, but not completely impossible; the Iridium satellite constellation has 72 operational satellites.

Is it accurate that Earth observation satellites will soon have daily 30 cm global coverage?

(for this claim, let's ignore the issue of clouds)

gerrit
  • 17,636
  • 17
  • 84
  • 137
  • 2
    Can we find a link to the claim? This sounds like exactly the sort of claim where the details matter. – Oddthinking Nov 02 '18 at 00:15
  • @Oddthinking Unfortunately I'm not sure if it's possible to find a transcript of the exact wording on the details for a TV news broadcast, nor is it available online (even within the UK). I've [asked on meta](https://skeptics.meta.stackexchange.com/q/4264/5337). – gerrit Nov 02 '18 at 10:07
  • "Soon" is a quite vague word. – Christian Nov 02 '18 at 10:22
  • @Christian I agree, I hope I can find the exact wording, but I don't know if I can. Maybe I'll send Jonathan Amos an email or a tweet. – gerrit Nov 02 '18 at 10:25
  • The BBC link is broken for me ("Sorry, this episode is not currently available"), so it's rather hard to tell what exactly was said. That said, I suspect that BBC misinterpreted the less than 1 day revisited time (at 40° latitude) at 1 meter resolution claimed for Worldview-4 as meaning global coverage. That is not what that means. Worldview-4 points its very narrow 1.2° FOV imager where paying customers want it pointed. If customer X wants to have the satellite view 20° off nadir at some time T, the satellite points that way briefly before time T and collects data for short amount of time. – David Hammen Nov 02 '18 at 12:33
  • The less than 1 day revisit time claimed by the [Satellite Imaging Corporation](https://www.satimagingcorp.com/satellite-sensors/geoeye-2/) does not mean what BBC appears to think it means, for two reasons. One is that revisit time is for reduced resolution, 1 meter rather than 31 cm. That corresponds to looking 56° off nadir. The other reason is that even then, the coverage is not global. The satellite has a very narrow 1.2° field of view rather than the 130° needed for global coverage. Ground controllers command the satellite to point in the direction paying customers pay it to point. – David Hammen Nov 02 '18 at 12:48
  • The 72 operational satellites claimed in the question is a bit off. I get 750 based on the [claimed capability](https://www.satimagingcorp.com/satellite-sensors/geoeye-2/) of collecting data for 680,000 km² per day. Worldview-4 spends most of its time transmitting previously collected data, slewing for the next paying customer, or doing nothing. The satellite collects data during a small fraction of the time. – David Hammen Nov 02 '18 at 13:08
  • @DavidHammen I thought 180 satellites because if it takes 6 months (~180) days to cover the entire globe then 180 satellites may be able to cover the entire globe in a day, but that may be an oversimplification. Your hypothesis on how the BBC reached this incorrect claim sounds plausible. – gerrit Nov 02 '18 at 14:05

1 Answers1

6

Given the technology of 2016 a defense company released the SpaceView 110 for non-Defense customers:

The SpaceView 110 camera magnifies its subjects with a mirror-based design more like a big telescope than a telephoto lens with lots of glass elements. It's precise enough to capture details measuring 0.31m -- about a foot across -- from the satellite's orbit 617 kilometers above the Earth's surface.

It has been reported that the limit of the resolution those satellites can provide is due to law that limits the resolution of commercial imaginary instead of being due to technological limitations.

According to Wikipedia the WorldView-4 satellite by Lockheed Martin has a mass of 2485 kg. One Falcon Heavy launch can carry 63,800 kg - more than twenty times the mass of a WorldView-4 satellite - to orbit for $90 million. However, as of 2012 the cost of the satellite itself was $835 million.

That's likely too high a price to put up hundreds of them, but they are going to be cheaper when produced at scale and Moore's law helps to drive the cost down as well. In 2017 the commercial satellite market was worth $3.061 billion.

EarthNow is a civilian project for real-time whole earth imagining that says this about their satellites' resolution:

The native video resolution, combined with image enhancement techniques, is designed to enable event monitoring and tracking applications consistent with existing and future customer requirements.

It's not clear whether that's in the realm of a 31 cm resolution but it illustrates that it's possible for a startup to track a lot.

Traditionally, the military spends a lot more on surveillance satellites and isn't keen on the public knowing about how good their surveillance capabilities are. NASA for example reuses old military tech for their earth survey satellites.

Given that aerospace is one of the areas of focus for China 2025, China is likely planning to put up a lot of spy satellites as well. China's space program is also more military lead then NASA in the US.

Given that there's huge military and intelligence value in knowing exactly what happens in another country, it's likely that both the US and Chinese military will spend significant resources upgrading their surveillance capabilities.

High resolution spy satellites aren't fixed to a geostationary orbit; a satellite that monitors North Korea at one time in the day might monitor New York at another time of the day. As a result a satellite program that's targeted to spy on countries like North Korea and Iran in high resolution will also be able to look at different countries.

It's unclear at which exact year the surveillance capabilities that match the BBC speaks about will come online, but "soon" is in itself an unclear word and it seems like it's only a matter of years until the those satellites fly over us. Societally, that makes now a good time to think about how to legislate the space. We will need to pass laws to protect our privacy in the face of EarthNow which on its homepage says it's willing to respect the privacy concerns of individual jurisdictions.

Christian
  • 33,271
  • 15
  • 112
  • 266
  • Based on the claimed capability of collecting 680,000 km² of data per day, a constellation of 750 such satellites is needed rather than 72. That multiplies your costs by a factor of ten. In addition, Moore's Law doesn't quite apply to satellites. To illustrate, the costs for Worldview-2, Worldview-3, and Worldview-4 were respectively 400 million US$, 650 million, and 835 million. The costs are going up, not down. What Moore's Law does for satellites is to improve capabilities but typically at increased costs. – David Hammen Nov 02 '18 at 13:19
  • 1
    @DavidHammen : Neither of those was mass produced. When you have better technology you have the choice to either increase capability or decrease costs. It's possible the the US military is unable to go for reduced costs due to their structural dysfunction in being able to budget, but that's unlikely to be true for either EarthNow nor the Chinese. – Christian Nov 02 '18 at 17:19