Profile picture
Alex Parker @Alex_Parker
, 54 tweets, 13 min read Read on Twitter
In just a few hours I will depart for Maryland for New Horizons' New Years flyby of the Kuiper Belt Object (486958) 2014 MU69. Before I go, I thought I would re-tell some of the stories about how we came to know about this little world.
I wrote a piece for NASA that tells some of these stories, and you can read it here:…
I'll be making some new illustrations as I go along here, so bear with me.
At launch, there were no known targets that New Horizons could reach after its flyby with Pluto. This is due in part to the fact that Pluto has been slowly crossing the plane of the Milky Way since before we knew the Kuiper Belt existed.
The Milky Way's crowded starfields confound searches for faint Kuiper Belt Objects in the foreground. This is what the known Kuiper Belt looked like on the day that New Horizons launched. See that gap around Pluto? That's the absence of discoveries caused by the Milky Way.
Given what was known about the Kuiper Belt at the time, though, it was clear that there should be _something_ close enough to New Horizons' trajectory for it to reach using its remaining on-board propellant. We'd just need to find it ourselves.
This wasn't going to be easy. We'd need to search very faint. We'd need to search over a fairly large area of sky. And that area of sky was just about the worst possible area imaginable.
We started using ground-based observatories with the best combinations of large aperture and large field of view. Here are a few of those, with mirror size (compared to the human figure) and FOV (compared to the Moon).
And what did we see? Here's a raw image from one of my first Magellan observing runs.
Fortunately, I am pretty good at erasing stars. Over the next few years we discovered about 50 new Kuiper Belt Objects in the patch of sky that New Horizons was flying into. Here's the known Kuiper Belt in mid-2014.
See that narrow streak of new objects radiating outward from Pluto? Those are the worlds we discovered.

Yet even so, none of them were within reach of New Horizons.
We did, however, manage to find a couple of weird things. One of those was the really high inclination Neptune Trojan 2011 HM102. I wrote about it here:…
2011 HM102 was also a fun discovery because it was one that a number of citizen scientists independently discovered through the Ice Hunters project. The discovery paper includes a full appendix of their names as co-authors.…
By early 2014 we were running out of time. If we were to find a post-Pluto target for New Horizons, we needed a finite about of time to refine its orbit sufficiently well to design and execute a precise engine burn to set the spacecraft on course.
That's when we turned to Hubble. The super-sharp images it delivers would reduce the TOO MANY STARS problem, and its sensitivity would allow us to see Kuiper Belt Objects far fainter than any ground-based observatory.
The challenge with Hubble for this kind of search is that it has a very narrow field of view, so we would need to tile many, many telescope pointings to cover an area of sky large enough to produce good odds of discovering even one targetable Kuiper Belt Object.
This means our survey was a huge ask, and somewhat risky. To help mitigate that risk, we were given a task to prove that (a) our models of the Kuiper Belt were good, and (b) we knew what we were doing with this data.
We were allocated a smaller pilot project, and we had two weeks to process that data, make any discoveries, and prove that we were discovering new Kuiper Belt Objects at the rate that we expected.
This was ... a lot of work.
We started getting data. Here's one of the first images that Hubble sent back to us. Still a lot of stars ... but you can see dark sky in between them. We assembled in Boulder and started working round-the-clock to develop the new software needed to turn these images into KBOs.
This is also when @AscendingNode and @AmandaZangari joined the search effort. Zero chance we'd have gotten this done without them.
So what did this search look like? Well, basically, we were looking for a tiny smudge amidst a sea of stars and cosmic rays. See that gif at the top of the thread showing "PT1" (which later became 2014 MU69)? Here it is with the cosmic rays intact.
For most astronomical targets, you can stack multiple images and the cosmic rays go away because they occur at random locations and times. Stars and galaxies don't move, so they remain. But solar system objects move, so a naive stack would erase them just as well as a cosmic ray.
So how did we find our faint moving smudges in a sea of brighter random smudges all varying with time? We used a method called 'shift-and-stack' or 'digital tracking.'
Basically, we say "if a KBO was moving on a particular orbit, it would move in this way." We would then shift the images to account for this motion and _then_ stack them. This removes cosmic rays, stars, and galaxies, but leaves behind any objects moving on that proposed orbit.
But ... there are an infinite number of possible orbits that KBOs could occupy. We can't search them all. Fortunately, as a grad student I had written a paper on how to choose an optimum set of these tracking orbits.…
By using my methods, we could usually get the number of digital tracking orbits we used down to about ~20. Sphere packing to the rescue!
And it worked! After a ton of work from the team implementing everything we’d learned from the ground based searches, it was @plutoflag who first email the team saying that one of the smudges looked real.
After that we quickly found a second KBO, and we were soon able to prove that the survey was working as expected. That’s when the data floodgates really opened.
You might imagine this was all being done on giant workstations with huge monitors. Welp, this was my entire workstation for the pilot program.
In the end, we had three completely independent data analysis pipelines running in parallel to ensure that we would miss _nothing_ in the data. This was all time-critical, because if we discovered a potential KBO we only had a short window of time to confirm that it was real.
The uncertainty in the motion of any new discovery was enough that follow-up with Hubble would become harder and harder the longer we took to take to schedule the follow-up observations.
In all, we turned up five new ultra-faint Kuiper Belt Objects in the search. There was possible a sixth that my pipeline alone turned up, but we were unable to confirm it in follow-up data.
We tracked them all so I could accurately measure their orbits. I ruled some out as potential targets right away, with orbits too far from New Horizons' trajectory. But a few remained as potential candidates.
And then, finally, on August 22, 2014, I got to send the fourth-best email I have ever sent. We had a confirmed targetable Kuiper Belt Object.
There was much rejoicing.
At this point, we knew a few things about 2014 MU69. We knew how bright it was, we knew the orbit it was on, and we knew that we could reach it.
2014 MU69's orbit put it squarely in the midst of a sub-component of the Kuiper Belt called the Cold Classical region (shown as orange points here). This is more or less what we expected, as it is the most densely-populated part of the Kuiper Belt along New Horizons' path.
You can see how different 2014 MU69's orbit (orange path) is from Pluto's (white path). 2014 MU69 has an almost-circular orbit that lies in the same plane as the major planets.
This orbit is part of what makes 2014 MU69 such an interesting target. A number of lines of evidence point to the Cold Classical Kuiper Belt being something of a relic, leftover from the era of planet formation.
Every other population of small bodies (like the asteroid belt) have been heavily stirred up by interactions with the planets since their formation. Not so for the Cold Classicals, it seems. They've been trapped, far from the Sun and the giant planets, for 4 billion years.
Our hope is that this flyby will give us a glimpse into what the first generation of planetesimals - the building blocks of the planets - looked like.
We got our first taste in the summer of 2017 when 2014 MU69 passed in front of a background star from Earth's perspective. By measuring the duration that the star 'blinked out' from different locations, we could map the shape of 2014 MU69's shadow.
While we can only get a snapshot of its cross section, it was pretty clearly not a sphere. Two sort-of-spheres might work okay, though.
I should point out that the gif illustration of the stars blinking out from 2014 MU69 passing in front of them, while highly idealized, is rendered in almost real-time. Occultations happen FAST.
A huge number of people contributed to the occultation measurements, and just placing the telescopes underneath the shadow of a city-sized object 4 billion miles away is an enormous feat of precision prediction.
These predictions were made by @AscendingNode and @plutoflag. I sat on the sidelines and yelled statistics jargon at them, too. None of it would have been possible without the ultra-accurate stellar positions measured by @ESAGaia.
If you're interested in the nuts-and-bolts of how we took the faintest Kuiper Belt Object ever tracked for more than a year and turned out ultra-precise spacecraft navigation data and occultation predictions, @AscendingNode has you covered:…
So as of right now, we know how bright 2014 MU69 is, we know the orbit it's on, we're on our way there, and we know that it's non-spherical and (possibly) binary in nature. Also, there doesn't seem to be any dust or debris in our way.
That's about all we'll know for the next few days.
But on January 1, 2019, New Horizons will fly by and we'll get to see 2014 MU69 up close. It will be the most distant, most ancient world we've ever visited, and we didn't even know it existed until four and a half years ago.
I have no idea what we'll see when we get there. There's no precedent to guide my expectations.
We've traveled four billion miles to go four billion years back in time, and I hope you'll come with us.
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to Alex Parker
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member and get exclusive features!

Premium member ($30.00/year)

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!