Hacker News Reader

Comment by jmpman | Hacker News Reader

How much Mercury does satellite propellant contribute to global airborne mercury, compared to coal fired power plant emissions? My first thought is that this is about 7 orders of magnitude less, and thus a ridiculous thing for these technocrats to focus on… but I’m happy to be educated. Or is there some other concern about mercury in satellite propellant?

View on HN

With the satellite megaconstellations being launched it seems important to ban it before it becomes common. Mercury-based fuels are cheaper than alternatives.

Coal as being burned for electricity has an industry standard of scrubbing 90% of mercury emissions, but is moving towards 99% using newer technologies.

So, it will be four orders of magnitude more for coal than for mega constellations. Seven minus one order of magnitude for better scrubbing and two orders of magnitude for the size of the constellations. Still a ridiculous thing to focus on.

If they would focus on reducing atmospheric mercury and direct their efforts against all offenders, I would completely agree. But singling out satellites or deliberately ignoring coal hints that they have a different agenda, and "atmospheric mercury" is their parallel construction.

> a different agenda

I'll bite: fossil fuel is running the show?

That would seem an obvious agenda, possibly. If you've ever seen the slander against electric vehicles or Tesla, you'll believe that the fossil fuel companies are promoting an agenda.

It's a smaller problem, but also easier to solve. The costs of a few suppliers changing their satellite propellants are orders of magnitude less than ending global coal usage (though both would be nice, of course).

Efforts are being directed against all offenders, that's why coal is moving to 99% scrubbing instead of 90% as mentioned up thread. As a species and civilization I think we can manage 'focusing' on more than one thing at a time.

Apart from anything else we should at least be consistent, right? If Mercury can reasonably be avoided in any form of polluting usage, it should be.

  > coal is moving to 99% scrubbing instead of 90%

Yes, but that resolves only one order of magnitude difference between the coal plant and the satellite constellations. Another two orders of magnitude are accounted for in the increasing size of the megaconstellations.

So even after those issues are addressed, earth-based coal is four orders of magnitude more problematic. Sure, focus on multiple targets. But don't pick a target 1000 times less significant to "diversify focus".

I think even if 1,000 people are getting cancer a year from a particular cause, that doesn't mean we can ignore this other cause that's only killing 1 person a year.

Should we also exclude small coal plants that only emit <1% of global emissions? What other cost effectively preventable emissions of mercury should we allow? Where do we draw the line? We should be consistent, especially since the costs of compliance don't seem to be all that great.

I hate to belittle a point, but this group is addressing the cause that is metaphorically killing 1 person per year while ignoring the cause that is killing 1,000 people per year.

In fact, I would go so far as to say that this cause is deflecting energy from any anti-coal movement by offering an new, shiny, space-age target for activists to concentrate on.

I have already addressed that point in this thread. The emissions from coal plants are being addressed with a 10x reduction already in the works. It is not being ignored, at all.

On the contrary, every effort to minimise mercury emissions and highlight their danger is worthwhile and raises awareness. The only reason we are talking about coal emissions now is because of this effort. If anything, consciously choosing not to address this specific issue and applying mercury emissions controls inconsistently would weaken the case for cracking down on other sources of emissions. It would smack of exceptionalism and preferential treatment. If you can carve out an exception for this source, why not others? Why do high tech rich world projects like a satellite fleet get a pass, but not a specific coal plant in a third world country with the same emissions?

  > I have already addressed that point in this thread. The emissions from coal
  > plants are being addressed with a 10x reduction already in the works.

What about the remaining four orders of magnitude? I've already factored in the 90% -> 99% reduction in my comments.

This is why I feel that such "feel good" efforts, such as focusing on the space industry, actually distract from the coal burning issue. You feel that things are being done to rectify coal, and I'm demonstrating that a 90% reduction in emissions (from 90% to 99% scrubbed) still pollutes four orders of magnitude (10,000 time more, though I accidentally said 1,000 more previously) than the proposed satellite constellations.

I have already addressed the question of orders of magnitude but I will try to be more explicit. The emissions from a single specific coal plant is a tiny fraction of total emissions. The emissions from a satellite fleet is a similarly tiny fraction of total emissions. Why should we crack down on one, but not the other? Your argument is a slippery slope that can be applied to any individual source of mercury emissions in that way.

You see coal emissions and satellite emissions as fundamentally different for some reason, but from the perspective of protecting the environment they are not. They are all just mercury emissions.

The solution is not to slice up mercury emissions sources into categories and issue exemptions or special treatment. We should consistently crack down on all sources of emissions, unless there is an actual specific reason to exempt them.

Of course there's a scale below which chasing down a small source of emissions might not be worth it due to costs, but banning mercury use in satellite propulsion is not a very costly use to crack down on. In fact it's already been done.

I appreciate that we both find the issue so passionate.

> What about the remaining four orders of magnitude? I've already factored in the 90% -> 99% reduction in my comments.

This thread started with a statement that satellite thrusters are at "first thought ... about 7 orders of magnitude less" than coal power plants, but where did that number come from? Others are noting that some proposed constellation would have emitted more mercury than the US, which contradicts that thought:

https://news.ycombinator.com/item?id=30841820:

> The US emits 52 tonnes/year into the air, apparently. The article claims _hundreds_ of tonnes for one constellation. Which does seem concerning if correct.

https://news.ycombinator.com/item?id=30840666:

> Apollo Constellation Engine designed their propulsion systems for mega constellations to use mercury, which is what got PEER involved in this. Popular Mechanics also reported on their plans. Even if Apollo changed their minds since, that's a pretty significant 'near miss'.

> https://ecfsapi.fcc.gov/file/111998371507/11_19_18_FCC_compl...

> https://www.popularmechanics.com/space/rockets/a25242578/apo...

What is your argument exactly? That as long as coal power plants emit airborne mercury, then anyone else can emit mercury as well as long as they emit less? This is just a form of whataboutism. We don't have ready, scalable alternatives to coal power at the moment, or don't have the will / money to bring alternatives into service. We do have alternatives to mercury-based space fuels.

It is simply easier to ban mercury propellants than it is to ban coal plant mercury emissions. A lot more people are dependent on the coal plants so the time and energy it takes to make changes there is just way bigger. Of course work is being done there as well, it just takes way longer.

The megaconstellations use reaction mass wheels and hall effect thrusters (Krypton in Starlink). I'm not sure any use mercury as a fuel.

You must have missed the "before it becomes common" part of the post you've replied to. No one is suggesting it's being used a lot. The ban is to stop it before it does.

But it's not even in use now, when, if it could be used as a cheaper fuel, it would.

Becoming common still needs for there to be some usage, even before it becomes common.

Apollo Constellation Engine designed their propulsion systems for mega constellations to use mercury, which is what got PEER involved in this. Popular Mechanics also reported on their plans. Even if Apollo changed their minds since, that's a pretty significant 'near miss'.

https://ecfsapi.fcc.gov/file/111998371507/11_19_18_FCC_compl...

The article claims "one proposed communications network using the technology could result in hundreds of metric tons of additional mercury"

And searching for mercury released by coal burning comes up with 50 tons in the US, and around 250 tons/year worldwide from all sources.

I have a hard time believing that figure for the satellite, but if true, it's actually significant.

Edit: Seems like an ion engine uses around 1 ton of propellant. So, yah, I guess banning mercury for this, really is significant.

Maybe mercury is different (though I doubt it), but current electric propulsion sats use nothing like that much mass. A Boeing 702 HP (a big GEO comsat) Xenon propulsion system has a capacity for 266kg of Xenon; uses about 5kg per year for station keeping. Constellation sats are usually smaller; a Starlink sat is about 260kg total. Dunno how much krypton they carry, but probably in the low 10s of kg each.

I got it from Wikipedia: "The Dawn spacecraft broke the record, with a velocity change of 11.5 km/s (41,000 km/h), though it was only half as efficient, requiring 425 kg (937 lb) of xenon."

I rounded up to one ton, since we're only trying to figure out order of magnitude.

Dawn is an interplanetary probe, though, and requires orders of magnitude more fuel than satellite constellations. For a more realistic figure, the Starlink satellites are fueled with about 5 kg of krypton.

Starlink alone has launched more than 2,000 satellites already and plans a total of at least 12,000. That's a lot of fuel - and that's just one constellation.

> That's a lot of fuel - and that's just one constellation.

Is that a lot of fuel? 5kg times 12000 is about 60 tonne. That's the weight of two big truck approximately. In which sense is that a lot of fuel?

And besides krypton is a noble gas, it doesn't react with anything.

60 tons of Krypton, eh. 60 tons of Mercury is huge.

I agree with you.

I think the concern was that some organizations were considering big low-altitude constellations. Once you go below 400km or so the drag increases quite a bit so the idea would be to use near-constant thrust similar to GOCE or SLATS.

And there was at least one vendor (Apollo Fusion) proposing a mercury thruster, although I think they have backed off on that. It would have had about 3x the impulse per volume and 2x the impulse per mass of xenon.

I think banning it now is wise, if any small operators had started using it it would have been hard to block it for a megaconstellation and it was very clear that anything release in VLEO would be coming back into the atmosphere. Right now there are no users so you aren't really penalizing anyone.

It mentions communications _networks_. If you have 10,000 satellites with 10kg propellant each, there's your hundred tonnes of propellant.

Coal plants are emitting a lot of unhealthy stuff.

The coal plant Jänschwalde in Germany emitted 672kg mercury, 1780kg lead and 208 kg arsenic in 2017 according to the European Pollutant Release and Transfer Register. https://de.wikipedia.org/wiki/Kraftwerk_J%C3%A4nschwalde#Emi...

This is the worst coal plant in Germany, but the others are not so much better.

The US emits 52 tonnes/year into the air, apparently. The article claims _hundreds_ of tonnes for one constellation. Which does seem concerning if correct.

Sounds pretty low actually, I'm sure it's much much more. 52 tons is more in line with the amount of gold in that coal. And hundreds of tons for one constellation? How do they get it up there in the first place?

These are big constellations of LEO satellites. Starlink, for instance, has 1500 satellites today, with a goal of 30,000. If we assume 10kg of propellant per satellite, then that's 300 tonnes for the finished constellation (Starlink satellites use a noble gas, I believe, but anyone using mercury for a similar scheme would hit that).

The whole _world_'s atmospheric emissions only seem to be about 250 tonnes per year.

I'm not sure. Wankawilca, the world's great mercury mine, and the most valuable mine in the world for centuries, produced mercury on a similar scale as America produced silver. They produced so much silver! I'm reading California had 45,000 tons of mercury not recovered, used in placer mining from 1848 til today. Yeah, no, it's as common as silver in the earth's crust, and it still gets used for artisanal gold mining, the only figure that's not way off is the amount the spaceships use.

There's a lot of historical falsifications in mining, because mining produces silver and gold, which is money, and generally people lie about how much money they have, telling a different lie to every person.

So Atahualpa's gold, there's no idea how much it was, and further they don't talk about how hard it was to get people not to constantly steal from that room, like what? You think a Spanish soldier is going to just stare at 100 times his weight in gold and be like, I can't wait til King Charles gets his share so I can have mine. You think those guys didn't take bribes for themselves, like OK we'll say you gave all your gold to the gold room, but you get to hang on to this much, and I get this much, nobody's the wiser. It must have been 1000 years of gold production of the Andes. That was el Dorado, they found it right away. And sources tell totally different amounts, some say 200 tons, which is stupid, well do you think they followed GAAP accounting standards when they were eyeballing the gold? How could they measure it? The whole thing was an extortion operation! Obviously the Spanish Conquistadors couldn't let the future know they were torturing the king...publicly? Yeah publicly. And tons of others too. And mining is always like that, it frequently involves slavery too, in complex ways. Because those metals are money, so it's controversial, and private. GM can say how many cars are on the road, but gold mines? Good luck! And I recently dealt with a gold mine someone I know invested in that failed despite excellent deposits because they said the miners were stealing from them--of course, the owners were stealing too. Coming full circle, that's why mercury was so important to Spain, because controlling mercury was the only good way they had of taxing silver and gold mining.

But I must also admit my ignorance presently, I don't know how to find sources that I would find satisfying.

To be clear, the world's annual production of mercury is far more than 250t/year; it's just that most of it doesn't end up in the atmosphere. The mercury that _does_ end up in the atmosphere is largely incidental, mostly due to coal burning.

The concern I'd have is that a launch failure could deposit all of the mercury on board in a small area near the launch site. Even if the amount involved isn't significant on a global scale, it would still be enough to have a dramatic and lasting local effect, e.g. on fisheries under the flight path.

“The potential release of mercury into the upper atmosphere by the firing of mercury-fueled thrusters could almost double the total atmospheric mercury emissions the U.S. is currently responsible for“

Thanks for HN educating me. This would become a significant source of modern US environmental mercury contamination. However, if you summed up all the coal fired power plant mercury emissions over time, this still seems insignificant. It’s not clear how long the mercury cycle takes for mercury to get locked into ocean sediment (effectively out of the food web). If it’s on the order of thousands of years, then satellite propellants are unlikely to be major contributors.

But the scale is surprising, and I would congratulate the FCC for pushing this restriction.

The way to prevent such 'unevenness' in rules intended to help the environment is to make them taxes and caps on emissions, rather than bans.

If for example the US were to decide we need to halve mercury emissions every 5 years for the next 50 years (a 99.9% reduction), then we would implement a system where anyone emitting mercury needed to purchase the 'right' to pollute by auction. Each year the available quota would be reduced.

The end result is the levels of pollutants are reduced in the most economically viable way.

Reply on news.ycombinator.com