End of 2025 Recap
2025 is suddenly over and I am behind on writing up some of the more noteworthy developments in my solar system object detection work this year. I should have posted all of these things earlier, but I'll highlight them now before the year ends at least.
ACX Grant
I was awarded $6,000 to fund my search for solar system objects! ACX grants are a microgrant program put on by Scott Alexander from Astral Codex Ten for charitable or scientific projects. Here are all the awardees for 2025.
The award will almost entirely go towards buying computers. I have already purchased one Mac Mini to replace the almost 6 year old laptop (!) which is currently running all of my Minor Planet Center ITF-ITF identification searches. In my testing I can search over far more hypotheses with the new Mac than I could with the old laptop and can find more objects as a result. This new setup for my ITF-ITF pipeline should be up and running shortly after the new year begins. And just today the MPC published the 2025 ITF-ITF linkage counts for contributors like myself here under ITF-ITF contributions.
I will also be buying a second Mac Mini with more CPU and memory than the first one in an attempt to be able to process at least a subset of the Rubin data that will come online when prompt products databases begin delivery in mid-2026 — that's the timeline as of December 2025 at least. I hope to be able to take on MBA and maybe partial TNO searches too. NEO and ISO searches will likely require more compute and/or clever algorithmic optimizations to be tractable. It's hard to tell what the first year of Rubin data will look like at this point as template fidelity is still a big unknown and will determine how many transient sources will need to be searched. In a sense, this new computer will be a yardstick for measuring how much compute I will truly need to do all the searching I want to be able to do with Rubin data. But it's a very good (and much appreciated) start for sure.
DES-DES Identifications
In addition to ITF-ITF identification, which I've been doing for a couple of years now, I started looking for DES-DES identifications this summer. An ITF-ITF identification is finding a handful of tracklets in the Isolated Tracklet File (ITF) that are all on the same orbit and formally declaring: "hey, these all belong collectively to one new object." A DES-DES identification is finding two (or more) already designated objects in the MPC's database of known objects that are actually the same object, not different ones. So instead of being designated as multiple separate objects they should all be designated as the same singular object.
When Rubin pushed a batch of newly identified objects in their First Look data for solar system objects to the MPC this past June, there were a handful of objects that were initially designated as newly discovered that later turned out to be existing objects. I think this was the first time I realized the scale and frequency at which this occurs. It's not a bad thing; the MPC has programmatic tools to rectify the situation if they or someone else finds these things.
I had a simple idea for how I might identify some of these DES-DES linkage opportunities. It's still evolving and I'll eventually write it up here in detail I think, but I'm not propagating the ephemeris to observation times and comparing positions on the sky. It's a clustering approach instead. Anyway, I set up a mini-pipeline for this that runs every morning and am happy to see that I'm making a few DES-DES identifications this way.
SAD Tracklet Poster
Going back to algorithmic optimizations, I presented a poster at the 2025 Rubin Community Workshop this summer. It's the same concept I showed in my Implied Orbits and SAD Tracklets post at the beginning of the year, but I think some of the details are maybe more clearly explained. This line of inquiry continues to be a productive one and has led to more insights which I hope to share in 2026. Here's the poster: Finding more NEOs with less compute using SAD tracklet rejection in HelioLinC.
Coming Soon: Vera C. Rubin's Legacy Survey of Space & Time
2026 starts tomorrow and Rubin's Legacy Survey of Space and Time may begin any day after the new year is under way. I've been looking forward to the start of LSST for years. The concept of LSST is what first planted the seed of trying to do object discovery myself way back in 2015 I think. But many of the researchers on the project have been working on it for literally decades (plural)! I think everyone in the planetary science world is excited to see what Rubin will discover. And just yesterday I found out that I am officially approved (s/Engebreth) to submit Rubin data to the Minor Planet Center. That's a necessary (but insufficient) condition to be able to use Rubin data to discover new objects. The other thing you need to do, of course, is actually pull something interesting out of the data.
Published: 12/31/2025