# Would it have been cheaper and/or faster to put a James Webb-like Space Telescope on a balloon instead of a rocket?

I've just noticed the following items:

The last one says:

Because far-infrared instruments need to be kept very cold, many missions carry liquid helium to cool them. ASTHROS will instead rely on a cryocooler, which uses electricity (supplied by ASTHROS’ solar panels) to keep the superconducting detectors close to minus 451.3 degrees Fahrenheit (minus 268.5 degrees Celsius) — a little above absolute zero, the coldest temperature matter can reach. The cryocooler weighs much less than the large liquid helium container that ASTHROS would need to keep its instrument cold for the entire mission. That means the payload is considerably lighter and the mission’s lifetime is no longer limited by how much liquid helium is on board.

The team expects the balloon will complete two or three loops around the South Pole in about 21 to 28 days, carried by prevailing stratospheric winds. Once the science mission is complete, operators will send flight termination commands that separate the gondola, which is connected to a parachute, from the balloon. The parachute returns the gondola to the ground so that the telescope can be recovered and refurbished to fly again.

Question: The JWST is absurdly late and over budget. Hypothetically, with hindsight, would it have been a better idea to put a JWST-like instrument on a balloon in a similar fashion to ASTHROS? There would have been a huge savings in R&D because every month the system could be serviced, so all the work and time and mass and energy making it space-worthy and ultra-reliable could be saved, and every month the instruments could be swapped out, repaired, improved, etc. rather than the final instrumentation being built once and sent to Sun-Earth L2 forever. Even Hubble benefitted from several swap-outs of equipment.

• At this point anything would have been faster and cheaper including building a second solar system to put it in. – Organic Marble Jul 26 at 13:49
• How far up would you like to go? A balloon needs atmosphere to push it up. – Thorbjørn Ravn Andersen Jul 27 at 7:47
• This seems to be a proposed follow-on from ASTHROS, using four balloons - note that they envisage ~ one flight per package per two years. – Andrew Jul 27 at 9:20
• @uhoh done! with some notes on the practicalities of doing it from the Pole :-) – Andrew Jul 27 at 11:52
• Wait - surely this should be on the Astronomy site ??? – Fattie Jul 27 at 16:38

No, I don't believe so. The reason space telescopes do well is that there's no atmosphere limiting the optical performance of the device. A telescope on a balloon is not anywhere near above the atmosphere. It's above a lot of the water in the atmosphere, which is why IR things can be better there, but there is still turbulence above it which will limit its performance.

Further, ASTHROS and JWST are not comparable instruments: ASTHROS works at frequencies in the far infrared, often defined as between $$25\,\mathrm{\mu m}$$ and $$350\,\mathrm{\mu m}$$, while JWST works in the visible and near infrared, from $$0.6\,\mathrm{\mu m}$$ to $$28\,\mathrm{\mu m}$$. ASTHROS is also tiny compared to JWST: its mirror is $$2.5\,\mathrm{m}$$ (about the size of Hubble's), while JWST's is $$6.5\,\mathrm{m}$$. If we take the extreme long end of JWST's wavelength sensitivity, where it overlaps with ASTHROS, it will have resolution about $$2.6$$ times as good. At the short end of its sensitivity range its resolution will be more than a hundred times better. And it's in space, so it may well be able to get really close to its theoretical limit.

JWST may be absurdly late and absurdly over-budget, but I don't think telescopes hung from balloons are competing with what it will be able to do.

However there is an important caveat to this answer: whenever someone says 'obviously such-and-such a thing is not possible' it turns out that astronomers have not only worked out how to do it, but are doing it and in fact have moved on to some even more absurd-sounding idea. So, I don't know, perhaps people are even now working out how to hang an optical interferometer from multiple balloons. Astronomers do amazing things.

• Feel free to post a similar answer to How does NASA's ASTHROS stratospheric telescope compare to its James Webb space telescope? in Astronomy SE. I think you've pretty much nailed this one already! – uhoh Jul 26 at 13:55
• @uhoh: I think it would be better if a proper astronomy person were to (and I'm hoping someone will as I'd like to know). My answer is basically full of what I'd call 'theoretical physicist arrogance' – making big sweeping claims based on pencil-and-paper arguments – and I'm not sure it's really right. (In fact while writing this comment I toned it down a little so it's less arrogant-sounding.) – tfb Jul 26 at 14:03
• to double check, by "optical performance" do you mean that ASTHROS' working altitude there will still be a problem with astronomical seeing or will the be an opacity problem for some wavelengths? – uhoh Jul 26 at 14:08
• @uhoh: I think mostly the seeing, but probably some opacity. But this is the right at the heart of the problem: I don't really know enough to be sure. We need astronomers (which is why your other question is good). – tfb Jul 26 at 14:17
• now I can't stop thinking about balloon-borne interferometers :-) btw per your caveat; Wikipedia's Balloon-borne telescope links to Sunrise which had a "Correlating Wavefront Sensor" which is used for micro-steering but not wavefront correction per-se. – uhoh Jul 27 at 6:14

Logistically speaking, it might be more complex than it sounds. The bit that initially stuck out to me is that the South Pole launch is planned for December 2023. December means summer - it is likely that the weather window during which it is practical to fly this mission is quite narrow, only a few months. Outside of that period, recovery becomes substantially more difficult.

This proposal (for a four-telescope version of ASTHROS) proposes flying each telescope package once every two years, which suggests a year for reconditioning and maintenance between launch seasons. Even if you did this - definitely pocket change by JWST standards! - you would be still be constrained by the operating seasons. You could only do observations for a few months in each year, and only of targets visible from the southern hemisphere in that period.

It might be possible to do flights around the north pole as well as the south - which would open up a few extra months, and also northern-hemisphere targets. But the Arctic has (I think?) less reliable weather patterns, and a greater chance of losing your instrument package by having it drop into the ocean.

You would still end up with a mission that might only be able to observe during parts of the year (I suspect weather conditions around the equinoxes might rule out both north and south), and would probably not give you full coverage of the sky.

If you could get a JWST-class telescope under a balloon for a twentieth or a hundredth of the price, which may or may not be possible, then it might well be a reasonable trade-off to have these limitations as well. But it would still be a trade-off of cost-effectiveness against limits.

• Also, it's polar day at this time of the year, which might also give more problems? – Paŭlo Ebermann Jul 27 at 21:59
• @PaŭloEbermann yes, that's a good point! You'd always be observing during illuminated (or at best twilight) periods. I don't know enough to say how much of an issue that could be, but it's certainly an added complication. – Andrew Jul 28 at 11:50

A couple more things to consider (that I haven't seen in the several existing answers) about having a James-Webb class observatory in the upper Earth atmosphere instead of at Sun-Earth L2:

• You've significantly degraded your available fields of view compared to Sun-Earth L2. Not only is your "below" completely occupied by Earth, but "above" has the moon and ever-increasing numbers of satellites to plan observations around. Not necessarily show-stopping, just a limitation to work around (and likely mitigated if you put the telescope at high latitudes, which appears to be the plan).

• You're above most weather, but not all weather. So long as the balloon is flying, turbulence has the opportunity to vibrate your mirrors and distort your seeing. This likely also limits the telescope's ability to track a given distant star precisely; surely you're talking about some kind of active steering of the telescope.

• These are all excellent points! Sunrise did indeed have active steering which tilted a secondary mirror based on a fancy wavefront sensor for example. On any given day JWST has a roughly 2𝜋 "sky" due to tilt limits of its sunshield, but over the course of a year that covers the whole celestial sphere, whereas a balloon at one pole has a lot less as you point out – uhoh Jul 27 at 22:19
• It might be somewhat better if both poles were alternately used but L2 does seem to have advantages. – uhoh Jul 27 at 22:20
• @uhoh yeah, if you start talking about fleets of stratospheric balloon telescopes I'm pretty interested. Their capabilities are likely to lag both ground and space observatories, but it certainly seems like they'd have a niche. – Erin Anne Jul 28 at 2:00

JWST is to be a 6.5-meter telescope, while ASTHROS is only 2.5 meters. That's a pretty big difference. On the other hand, perhaps you could spend half the cost of JWST and engineer a 6.5-meter balloon-hosted telescope, but I'm not sure.

• The question asks about a James Webb-like telescope but you've answered with a ASTHROS-like telescope. Considering that this balloon is already huge, I've seen nothing so far that says this is the limit and a larger diameter telescope couldn't be lofted. However you might want to look into the flatness requirements for the two mirrors, and their masses! – uhoh Jul 26 at 15:29

The place to start is by listing the science objectives of the mission. Balloon telescopes can be much less expensive than satellite borne ones, but the design of a satellite borne one allows a much longer life span, zero gravity to distort the telescope, less infrared heating from the earth, a wider field of view, and I'm sure many more things. You would need to design a completely different telescope from JWST to fly on a balloon. Could it do what JWST is planned to do? I strongly doubt it, but I frankly do not know.

• +1 My question begins with "Would it have been..." to allow for some 20/20 hindsight. Considering that JWST was supposed to launch in 2007 and in the following 100,000 hours it has done exactly zero, a balloon launched IR telescope that did anything would have done infinitely more than that. – uhoh Jul 28 at 4:10
• The problem is that JWST was designed around the fact that it would be in zero G and be a long duration mission. You can't just suspend that design from a balloon and have it work at all. If you redesign JWST to fly on a balloon you get a completely different instrument and completely different science. Then you might try to ask whether we should buy JWST or the number of balloon missions we could get for the same money. That can't be answered because we could buy so many balloon missions the output of the last ones is really hard to predict.... – Ross Millikan Jul 28 at 4:20
• I think we are left with deciding whether to buy each mission based on its price and expected output. For major observatories like JWST, that price comes with a risk that needs to be addressed. – Ross Millikan Jul 28 at 4:21

The JWST will be in a halo “orbit“ at the Earth-Moon L2 point, 930,000 miles from the Earth, and over four times as far away from us as the Moon is. It would need to be one impressive balloon to achieve that kind of altitude. And it has to be that far away so that its sunshade can deflect heat from both the Earth and Sun, which is necessary for an infrared telescope. At the L2 point they’re always in the same direction, which isn’t true for any closer location.

• And yet I give an example of a cryogenically cooled far infrared telescope hanging from a balloon in the question, and Wise/NEOWise's spectral response went out to 22 microns and was in LEO, so I think this answer needs some refinement. – uhoh Jul 27 at 15:59
• Wikipedia says: "Infrared telescopes may be ground-based, air-borne, or space telescopes. They contain an infrared camera with a special solid-state infrared detector which must be cooled to cryogenic temperatures." There are plenty of bolometric imaging far infrared telescopes on the ground in antarctica, including this one so I don't think the situation is so cut-and-dry. But +1 for taking the time to answer! – uhoh Jul 27 at 16:07

One point touched on but not expanded upon is high contrast imaging. That's less important when imaging a black hole or a nebula or galaxy because they don't change much and the image can be reconstructed.

Imaging a planet orbiting next to or crossing over a star requires very high contrast and that wouldn't be possible with an airborne telescope, doubly so when the atmosphere blocks some of the wavelengths that the telescope is designed to image.

Image reconstruction is also (I remember reading) easier with UV light and less successful with IR light, so out in space is simply better for specific things like exo-planets. Probably much better.

The James Webb Space Telescope is expected to be able to make an observation between 1/10 million and 1/100 million in variation. An atmospheric telescope, especially one where some of wavelengths it's designed to receive are partially blocked, would never do that.

https://www.nasa.gov/feature/goddard/2019/a-new-view-of-exoplanets-with-nasa-s-webb-telescope

Coronagraphs have something important in common with eclipses. During an eclipse, the Moon blocks the light of the Sun, allowing us to view stars that would normally be overwhelmed by the Sun’s glare. Astronomers took advantage of this during the 1919 eclipse, 100 years ago on May 29, in order to test Albert Einstein’s theory of general relativity. Similarly, a coronagraph acts as an “artificial eclipse” to block the light from a star, allowing planets that would otherwise be lost in the star’s glare to be seen.

“Most of the planets that we have detected so far are roughly 10,000 to 1 million times fainter than their host star,” explained Sasha Hinkley of the University of Exeter. Hinkley is the principal investigator on one of Webb’s first observation programs to study exoplanets and exoplanetary systems.

“There is, no doubt, a population of planets that are fainter than that, that have higher contrast ratios, and are possibly farther out from their stars,” Hinkley said. “With Webb, we will be able to see planets that are more like 10 million, or optimistically, 100 million times fainter.” To observe their targets, the team will use high-contrast imaging, which discerns this large difference in brightness between the planet and the star.

• This is a really important point, and I didn't know JWST even had a coronagraph, but now that you mention it of course it makes sense; I think there might even be a more favorable brightness ratio in thermal IR than in visible. But I'll have to think about that... or ask about it ;-) – uhoh Jul 29 at 10:07