Taking a break from galaxies and cosmology, I’d like to post a little praise of NASA for safely returning a piece of an asteroid to Earth.
One of the amazing things to me about astronomy & astrophysics is that we have learned how to decipher the composition of distant stars and gas clouds by observing their spectra. I worked on this early in my career and retain an interest in the cosmic abundance of the elements. Fun fact: though often overlooked because it is a boring noble gas that doesn’t bind chemically into any common molecules or minerals, neon is number 5 on the list of most common elements, which goes hydrogen, helium, oxygen, carbon, neon. Nitrogen is number 6 by number, but iron supplants it if we weight by mass – there are more nitrogen atoms by number in the sun but the iron weighs more because of the greater mass of each atom. The order of the first five remains the same by either accounting.
Amazing as it is that we can do this, it can only be accomplished by passive observation. What we’d really like to do is get samples of the remote universe to analyze in the laboratory where precision is much higher and we can better control for systematic effects. Of course we can’t travel to stars and nebulae that are many light-years distant, let alone return from there. But we can do it within the solar system, which is amazing enough. The Apollo astronauts brought back rocks from the moon that helped determine the age of the solar system (4.568 billion years, give or take a million), and the period of “late heavy bombardment” when most big lunar craters were formed – a mere 3.9 billion years ago. This in turn calibrates crater densities; counting craters on other solar system bodies lets us gauge the age of a surface. Lots of craters means old; few craters means something interesting had to happen to cover up all the craters that formed during heavy bombardment. It’s not like all those early meteoroids were dodging the Earth while hammering the moon; it’s just that the Earth has covered it up since.
One of the most interesting things scientifically are samples of pristine material – the stuff from which the solar system formed. The Earth is a remarkably active planet geologically, which means that its rocks are always getting remade by erosion, subduction, and volcanism. They’re about as far from pristine as a rock can get. The closest we expect we can get are the comets and asteroids orbiting safely away from the big planets that have a complex history of their own.
Hence the idea for a mission that could return a sample from a remote asteroid. This is what OSIRIS-REx has now accomplished. It is worth pausing to reflect what an amazing feat this is.
We’ve only had the capacity to launch things beyond the atmosphere of our planet for 66 years. Though satellite launches are now relatively common, the most frequent destination is low earth orbit. That’s only a couple thousand kilometers, which is about a third of an Earth radius, so still pretty close. It is a distance that planes traverse horizontally all the time, if only at an altitude of 10 km or so. It’s just not that far on an interplanetary scale.
Deep space missions that leave Earth’s gravity well are harder and much less common. Those that go out to an asteroid, grab a piece, and return are even harder. It’s one thing to shoot something off a rocket so hard it never comes back. It’s quite another to do that and then turn around and come back at a time and place of our choosing. That’s a remarkable feat of celestial navigation and rocket engineering. Oh, and pause on the way to graze an asteroid, grab a sample, and store it for safe return.
Safe is key here. If one wants a pristine sample of the early solar system, you not only need to go to deep space to collect it, but you have to keep it safe through the rigors of reentry, collect it, and get it to your lab unsullied by terrestrial contaminants. Lots that can go wrong. The spacecraft has to endure the heat of reentry, suffer no leaks, and land gently in a spot where the sample can be retrieved. This all went well for OSIRIS-REx. It doesn’t always work so well.
Genesis was another sample return mission. Launched in 2001, it collected particles from the solar wind – a good way to get a measure of the composition of the sun. It did this for several years before returning 19 years ago to the month, to the same landing area as OSIRIS-REx. As it happened, I had just flown to Tucson to observe at Kitt Peak, and found myself having breakfast in the La Quinta next to the airport before renting a car to drive up the mountain. The landing was on the TV there, so it was breakfast and a show.
Only the show didn’t go so well. A helicopter was supposed to snag the capsule as it drifted at the end of its parachute to ensure no contamination from the ground. Through some amazing camera work, they showed a fairly zoomed-in image of the return capsule as it hurtled from the sky. Spinning, spinning, spinning… it looked out of control. Shouldn’t the parachute have deployed by now? Maybe not – that’s often done at fairly low altitude where the air is thick enough to bite. So I watched, spinning, spinning, as seconds stretched into minutes, spinning, spinning, surely the parachute will deploy any moment now, spinning, spinning, any moment now, spinning, spinning, really, any moment now, spinning, spinning, SMACK! into the ground.

The parachute failed to deploy. Apparently Lockheed Martin installed it backwards, a mistake for which I’m sure they were well remunerated. This is but one of the hazards of space travel.
So it was with a little trepidation that I watched the return of OSIRIS-REx this morning. There was again some amazing camera work. First we saw the blaze of reentry, then after that faded the capsule itself emerged, becoming visible while still at high altitude. Spinning, spinning.
As I was watching on NASA TV, it was announced that the order to deploy the parachute had been issued. Spinning, spinning. Good. Spinning, spinning. No parachute. Was there a time delay on that order? Still seemed high to be deploying a chute, but it was hard to judge the altitude from watching a small spinning blob on TV. Spinning, spinning. I am old and jaded, so I didn’t feel nervous – yet. Spinning, spinning. Only a tiny bit of anxiety. Spinning, spinning. Then it was announced that the parachute was scheduled to deploy at 49 minutes past the hour – still two minutes away. Spinning, spinning. Then, at 48 minutes past the hour, the parachute deployed. I was so enthused to see it that I didn’t worry that it had come a bit early – better than too late! Apparently it deployed at an altitude of 20,000 feet when it wasn’t supposed to deploy until 5,000. So that went wrong, but only a tiny bit wrong – it came gently to rest on the ground near the edge of the target ellipse – i.e., within the error bars.

This time there was no unnecessarily elaborate plan to snag the capsule out of the air with a helicopter as there had been for Genesis. But a helicopter was used to transport the capsule, dangled from the end of a long rope, to a temporary clean room that had been set up nearby. From there it will be transported to the Astromaterials facility at the Johnson Space Center in Houston, where they have an office of Astromaterials Acquisition and Curation. Sounds very Indiana Jones in space.
Science to follow.
So many entries – what a gift!
Thanks for your ongoing commitment to great astrophysics communication.
Quick question that’s been on my mind and only vaguely relevant. With this uplifting post about astronomical experiment and space missions I’d love to ask – could someone propose an experiment where a satellite or probe might be launched carefully to probe EFE/Mond dynamics within our own solar system?
What kind of MOND dynamics could be explored, if any, if a small single object like a probe was launched and manuevered to have an acceleration in the very low regime where it might show a difference from Newtonian dynamics?
No need to be gentle.
Thanks,
GPB
This is indeed an obvious thing to do. The hang up is getting far enough out to detect interesting effects. One has to travel about 7000 AU – about a tenth of a light-year – to get to a0 away from the sun. Traveling that far takers a prohibitively long time – no one has the patience to PI a program that won’t return results within their lifetime. We’re a short-sighted species that way.
That said, one doesn’t have to get quite that far out to detect effects with a high precision experiment. It does have to be extremely high precision – the famous Pioneer anomaly, now attributed a slight asymmetry in radiative heat loss, was a tiny effect but also too large to be caused by MOND. So on has to do better to see effects even modestly far out.
Still, it should be possible – one just has to take care to make purely ballistic beacons whose trajectories can be tracked. I’d suggest packing a bunch of them into a payload on a rocket that blasts them as far out as fast as possible, and once they’re well past the Kuiper belt, explode them off in many directions so that there are many probes on unique trajectories.
Nature may provide such probes in the form of Sedna group dwarf planets. The orbits of these objects seem to show the effects of a quadrupole moment that is interpreted to indicate the existence of Planet 9. It is also consistent with the effects of the Galaxy on the outer solar system via MOND – see https://arxiv.org/abs/2304.00576. So IF the data for Sednoids persist in showing a quadrupole moment and IF Planet 9 is not discovered, then this would be the sort of detection of MOND that we’re talking about looking for with a satellite probe.
Solar System tests of MOND are considered in detail in the future tests section of my MOND review (Arxiv: 2110.06936). One just has to reach the Sun-Neptune saddle point.
Saddle-point tests are interesting but are specific to particular flavors of theories. TeVeS, for example, predicts large effects where the gradient of the potential disappears, so any saddle point provides a test, at least in principle. The region of large effects was of order one meter, so would be hard to hit. More generally, I don’t think this is not a generic prediction of all possible MOND theories. I went through this in detail over a decade ago; that’s all I recall offhand.
The saddle point is hundreds of miles wide for Neptune and takes perhaps a week to cross. It is quite a promising test of classical modified gravity theories that are supposed to work down to metre scales. I had a masters student do a project on this, which I can find if people want. Though the main points are explained in the review. I suggest looking at the future tests section towards the end of the review where a few pages are devoted to the saddle point test. It would not really work for the Earth-Sun case. But there are other planets.
Huh, why Neptune and not Jupiter?
The size of the saddle region, according to the paper mentioned above, is proportional to the cube of the distance and the square root of the mass. Neptune is 20 times lighter than Jupiter, but 6 times further away from the Sun, hence a better candidate (if I understand all this correctly).