Chinese city to launch man-made moon to light up skies | Hacker News Reader

Chinese city to launch man-made moon to light up skies

atimes.com

40 pointsdankohn159 comments

Relevant math here:

http://jxcrystals.com/publications/Mirrors_in_Dawn_Dusk_Orbi...

Money quote: "Applying this formula for a mirror in orbit at an altitude of 4200 km gives a sun spot diameter on earth of 42 km".

The report claims "The satellite would be able to light an area with a diameter of 10 to 80 kilometers, while the precise illumination range can be controlled within a few dozen meters", so either it's in a lower sun-synchronous orbit [1] (and maybe there is more than one) or it's a parabolic reflector [2] with controllable focal length, and there is no cause for concern. [3]

[1] https://en.wikipedia.org/wiki/Sun-synchronous_orbit

[2] https://en.wikipedia.org/wiki/Parabolic_reflector

[3] https://en.wikipedia.org/wiki/Sun_gun

Depending on the distance from Earth, it seems it could also act as a statite [1], which might allow for additional options in terms of illumination timeframes.

[1] https://en.m.wikipedia.org/wiki/Statite

And they're assuming $0.1 per kWh, which is ok for the US, but could mean faster ROI around the world.

https://www.ovoenergy.com/guides/energy-guides/average-elect...

https://www.npr.org/sections/money/2011/10/27/141766341/the-...

Colour me moonglow sceptical blue. At an altitude of 500km these 'moons' would cross the sky in a few minutes. Geosynchronous orbit is about 35,000km, but at that distance any artificial satellite would need to be vast to cast any significant light. Regular satellites at that distance aren't naked-eye visible.

This report is, at best, seriously garbled.

They can be visible its just relatively rare and only happens at certain times of the year depending on altitude when the solar panels line up. Flares like these are pretty much just a tiny accidental version of this already, something intentionally designed and controlled to produce them would in theory be pretty easy.

My main thought is how well will this deal with the thrust that large of a mirror will be producing. It will essentially be a solar sail. Maybe they plan to rotate it during the day to provide the thrust to put it back into position?

https://www.cloudynights.com/topic/530024-geostationary-sate...

Interesting thanks, I didn't realise they ever got that bright. I've also captured geosynchronous satellites when imaging the Orion Nebula, they show up as parallel tracks.

There is a big difference between producing a visible flare and actually providing illumination at that distance, that would require a huge surface. My best guess is that the original press release was talking about a paper project rather than anything there are any real plans to do.

There a large difference is but those flares are all produced by relatively small surfaces that aren't designed to reflect significant amounts of light (quite the opposite in fact) and a purpose built device.

> This report is, at best, seriously garbled.

It's just run-of-the-mill journalism. Just write what it says in the press release, don't ask any questions.

I am sure the astronomy community will be thrilled... Infinite light pollution.

I was thinking the same thing. I don't know how much light they will emit, but as someone who grew up in a village with very little light pollution; even something like a "satellite light" results in some pollution.

I used to be able to see many constellations with the naked eye, until they put up 3 lights that are about 20m tall. My astronomy interest has waned as I can't see much with the lights. Now cue the "moons"

I wonder what the impact will be on wildlife that depends on the hours of darkness?

If it works as intended it should be mostly focused over the city so the effects will be a lot lower than if they were illuminating a random patch of forest. I doubt they're particularly concerned with the environmental impacts though.

Interestingly, until now, "light pollution" meant artificial sources. (i.e. with different colour and spectrum signatures such as CFLs, LEDs, etc.) Florida Atlantic University studies the impact of light pollution on plants[1] and animals[2] but I suspect many studies only partially apply to something like what Chengdu is doing because significant solar (or lunar) light pollution hasn't been available to study.

Just a few concerns off the top of my head that I hope Chengdu will consider or at least study if it moves forward with this: circadian patterns, melatonin production, restful sleep, feeding patterns of nighttime predators (even in cities) including exhaustion of food sources (e.g. insects) that would naturally be difficult to find.

Well this will probably improve human and traffic safety, but seriously hurt astronomy and dark sky appreciation for those who live there. And it's a given that all-seeing surveillance footage resolution will be improved!

[1] http://cescos.fau.edu/observatory/lightpol-Plants.html

[2] http://cescos.fau.edu/observatory/lightpol-environ.html

Here's a closer-to-ground design from over a century ago: https://en.wikipedia.org/wiki/Moonlight_tower

And a reflective one: https://www.theatlantic.com/photo/2013/10/using-giant-mirror...

Every satellite I've seen crosses the sky in seconds. How does an orbiting mirror provide consistent illumination? Can it orbit as show as the rotation of the Earth and still maintain an orbit?

See below

GPS satellites are not in geosynchronous orbits. In fact, they would be close to useless if they were (all in one line orbiting over equator).

https://space.stackexchange.com/questions/10837/why-are-the-...

Why would a geosynchronous satellite have to be in one line over the equator? Why not a higher inclination?

If the orbit is inclined the satellite will no longer stay above a single point it will move north and south throughout the day. All orbits pass over the equator twice during each period.

That's true, but it doesn't answer my question.

Geosynchronous satellites can and do modify their inclinations and eccentricities to increase their coverage beyond one spot over the equator.

They can have inclination other than 0. If inclination is 0 thats a "geostationary" orbit, a special case of geosynchronous. In fact these are not stable and degrade into mere geosynchronous orbits without station keeping. In general though geostationary orbits are more useful since they are fixed competely in the sky. Ground antenna dont have to track the satellite as it moves north/south in the sky.

In this case a pair of satellites in a Tundra orbit [1] inclined to spend most of its time over this particular city seems like the right orbit for the job (as silly as it is to light a city by satellite), which is why I was curious to get an explanation behind the sweeping statement that geosynchronous (not geostationary in particular) orbits had to be in a line along the equator.

1. https://en.wikipedia.org/wiki/Tundra_orbit

Oh thats interesting. I hadn't heard of Tundra orbits. Wiki says 2 can provide continous coverage over an area. Wouldnt 1 be enough to act as a moon then? You only need it for half of the day and their period is 1 day right?

The problem with using just one is that geosynchronous orbital periods are one sidereal [1] day, not one solar day. So depending on the time of the year, the satellite would be in position in the wrong half of the day. On the other hand a satellite in Tundra orbit spends not half but more than half of its time in the designated area. Maybe a good compromise orbit could be found.

1. https://en.wikipedia.org/wiki/Sidereal_time

Most people use geosynchronous as a synonym for geostationary.

https://en.wikipedia.org/wiki/Geosynchronous_orbit Popularly or loosely, the term geosynchronous may be used to mean geostationary.[2] Specifically, geosynchronous Earth orbit (GEO) may be a synonym for geosynchronous equatorial orbit,[3] or geostationary Earth orbit.[4] Communications satellites are often given geostationary or close to geostationary orbits so that the satellite antennas that communicate with them do not have to move, but can be pointed permanently at the fixed location in the sky where the satellite appears.

Certainly the comment I responded to (which has been edited to be non-sensical since) meant geostationary. In any case, GPS are not geosynchronous in the general sense of the term, but rather semi-synchronous (period of half a sidereal day). These orbits were historically chosen for convenience, but the syncronicity is not at all a requirement for global positioning satellites: e.g Galileo, GLONASS.

All GPS satellites are in a geosynchronous orbit.

Citation needed?

You may find this helpful:

http://www.astronomy.ohio-state.edu/~pogge/Ast162/Unit5/gps....

The nominal GPS configuration consists of a network of 24 satellites in high orbits around the Earth, but up to 30 or so satellites may be on station at any given time. Each satellite in the GPS constellation orbits at an altitude of about 20,000 km from the ground, and has an orbital speed of about 14,000 km/hour (the orbital period is roughly 12 hours - contrary to popular belief, GPS satellites are not in geosynchronous or geostationary orbits). The satellite orbits are distributed so that at least 4 satellites are always visible from any point on the Earth at any given instant (with up to 12 visible at one time).

No, they are not.

Let's do some quick math.

Geosynch is 36,000 Km out.

The Moon's albedo is 0.12.

Let's assume we can get a mirror of albedo 0.96 -- this is a little better than polished silver. It's also conveniently 8 times as reflective as the moon.

The Moon's angular diameter is about 0.5 degrees. To match that angular diameter at 36,000 Km, we need a 5.4 Km wide object.

That doesn't sound like it's within the current state of the art. Maybe in another 10 years?

23 million square meters at 2 g/m^2 (aluminized mylar) is 46,000 Kg -- about two Falcon Heavy trips. That doesn't count any framework or booster or other infrastructure.

Not sure, but it seems to me that for a mirror, more than albedo matter, because mirrors have a specular reflection rather than a diffuse reflection.

Now, reflection won't be perfectly coliminated, but if you assume it is, then the mirror essentially becomes a piece of sun. Since the sun is about 400 000 times as bright as the moon, you'd need to scale the diameter of the sun by sqrt(400 000 / 8) =~ 223. Applying that scaling to your 5.4Km yields about 24m of diameter, which seems waaaay to small to me, so I am probably wrong.

That feels really

Not the sun, the sun's irradiance at the distance of the Earth-Moon system.

> The Moon's angular diameter is about 0.5 degrees. To match that angular diameter at 36,000 Km, we need a 5.4 Km wide object.

The article doesn’t say it’s going to be as big as the moon - it only says it’ll be 8x brighter. You’re also forgetting about the relative shapes of the objects: the moon, being round, reflects what light it does in all directions, while a designed mirror can be made to reflect it all in one specific direction.

All that said, I don’t think this will work, but not be of your math.

8x as much light as the moon is the same angular diameter and 8x more reflectance, or the same reflectance and 8x as much angular diameter. I suppose they could mean "8x as much reflectance but an unknown smaller insolation equivalent", but then they might as well say "we're putting a big spot light on a high-altitude balloon".

The relative sphericity of the moon is nearly immaterial here: when the moon is in full phase, do you see notable dimness around the edges where less sunlight is reflected away from you? Increasing the albedo to a near-perfect mirror cancels it all out.

Ooh. Radiation pressure. 23 million square meters * 8 microN/m^2 = 184 Newtons of thrust away from the Sun. Roughly 1/250 G of acceleration. That doesn't sound like much, but you need to counter it for stationkeeping. It's going to be a big chunk of the mass budget.

> That doesn't sound like it's within the current state of the art

> about two Falcon Heavy trips

This does sound like it's within the state of the art! The mirror doesn't have to be one solid piece, it can be assembled from many smaller-ish satellites.

Of course you have overhead, so ... 5 Falcon Heavy trips? This is still very fine, if you're willing to put in the money.

Yeah, but there's only one such "ring" around the equator and it's pretty crowded. https://en.wikipedia.org/wiki/Geostationary#Orbit_allocation

Geostationary orbit also requires the sattelite to take on an equatorial orbit. I doubt a geostationary orbit will be used in this case.

Is there any kind of funky thing they could do to have it rotate slow enough to take an entire night to cross the sky? something like a lagrange point perhaps (stationary relative to sun?)

The closer they are, the easier it is to illuminate. I believe that excludes most Lagrange points.

The Lagrange points are weird because they orbit two things at the same time. There should be more normal orbits somewhat lower than geostationary that cross slowly, but they would also be out of sight for a long time.

Then I suppose the idea would be to find an orbit where that long time is equal to one day - obviously, the length of one day on a given section of the planet keeps changing, so perhaps you only turn on the light after a certain amount of time or find some orbit that mimics whatever the sun is doing if thats at all possible

Geostationary orbits are definitely a thing. Haven't done the math but two should take care of it, one east and one west of the area you're trying to illuminate and one should come out of the umbra before the other one enters.

edit: One might also take care of it if it could be far enough east. Would need to do the trig though.

At the equator.

Yes it would be orbiting above the equator but at GEO it can easily see all of China which is all it need to do to reflect light from the sun to Chengdu.

I guess it would have to be placed in geosynchronous orbit. I didn't see any size comments, but that puppy would have to be quiet large I'd think. To light 14,300-square-meters from 35,786 km seems like you'd need a BIG reflector.

And, of course, Chengdu is actually 14,300 square KILOMETERS, not meters. (Its 11 million residents aren't quite that cramped).

That would be 119 m on each side... yeah, that's a little cramped.

I copied that totally not critically from the article. I'm beginning to believe that this article may be missing some important information.

Yes, a satellite can appear to cross the sky as slowly as you want it to if the altitude is correct. This is how geostationary satellites (like the GPS and many communications ones) work - they are at the right altitude so that they are orbiting the earth at the same speed it rotates.

GPS is not geostationary, it's in a MEO about 12k miles up (GEO is about 22k miles) and they orbit about twice a day.

Geosynchronous is the word you’re looking for.

Most communication satellites are in geosynchronous orbit.

Sounds good in theory but would be interesting to know how prevalent cloud cover is in the city.

Or smog

I wonder if there's more than meets the eye behind this?

"That's no moon...."

Anyone seen Goldeneye recently? Great movie.

Vegeta did it first.

Cf. Austin, Texas.

Reply on news.ycombinator.com