Main Engine Cut Off - T+142: T.S. Kelso, CelesTrak and AGI

Episode Date: December 16, 2019

Last week, I took a ride out to the AGI offices and sat down with Josh Poley and T.S. Kelso. We shot a handful of videos for AGI’s YouTube channel, the longest of which was this interview right here.I talked with T.S. Kelso about the history of CelesTrak.com and satellite tracking on the internet as a whole, as well as a few topics relevant to the modern day: satellite tracking and orbit reporting among operators, conjunction and collision monitoring, and space debris mitigation and management.Be sure to follow along with AGI’s channel as the rest of the videos go live throughout the week!This episode of Main Engine Cut Off is brought to you by 38 executive producers—Kris, Pat, Matt, Jorge, Brad, Ryan, Nadim, Peter, Donald, Lee, Chris, Warren, Bob, Russell, John, Moritz, Joel, Jan, David, Grant, Mike, David, Mints, Joonas, Robb, Tim Dodd the Everyday Astronaut, Frank, Julian and Lars from Agile Space, Tommy, Adam, Sam, and six anonymous—and 314 other supporters.TopicsAnalytical Graphics, Inc. (AGI)ComSpOC | AGIAnalytical Graphics, Inc. - YouTube35 Years of Tracking Satellites - Spacecast Ep25 - YouTubeT.S. Kelso (@TSKelso) | TwitterCelesTrakDr. T.S. Kelso, CelesTrakThe ShowLike the show? Support the show!Email your thoughts, comments, and questions to [e-mail] @WeHaveMECOListen to MECO HeadlinesJoin the Off-Nominal DiscordSubscribe on Apple Podcasts, Overcast, Pocket Casts, Spotify, Google Play, Stitcher, TuneIn or elsewhereSubscribe to the Main Engine Cut Off NewsletterBuy shirts and Rocket Socks from the Main Engine Cut Off ShopMusic by Max Justus

Hello and welcome to Main Engine Cutoff, I am Anthony Colangelo, back again with another special edition of MECO. Last week I got the chance to visit a pretty cool spot. So I took a drive out to Exton, Pennsylvania, which is just about an hour west of Philly. A little longer if you take the Schuylkill, which I unfortunately did. But that's some local trivia. And I visited a company called Analytical Graphics Inc., or AGI. Now, anyone out there that works in the space industry probably knows AGI. They make a suite of tools for operators in space,

people that are actually working on this stuff day in and day out. Some really cool stuff that they're working on over there. So I got to go out and hang out there for a day. So we got to visit the Commercial Space Operations Center, which is a really cool spot that they've got right in their office there, the comms box, as they call it. And what this is, is a place that tracks satellites, that keeps track of any conjunctions between satellites, things that might be dangerous, things that they might need to avoid. And they do this commercially. So people pay them money to provide them this service of space situational awareness. That's keeping track of where things are, any situations that might arise that you need to be aware of. So I guess it's SAS, Space Situational Awareness as a Service. That's a tech joke. Got to go in there, see all the different stuff that they're working on,

and get a tour of the really cool looking center there. And there was this screen on the side that kind of showed upcoming conjunctions. And it was just like constantly flipping through this screen of just showing, you know, here's things that are really close together in space, and they're only a couple kilometers apart, or they're one kilometer apart. Just some stuff that gave you a little bit of anxiety. And so I got a full tour there, and there was a lot to see. I'm hoping that maybe I can visit again soon, and we can do a whole show on ComSpot

because there was so much going on there. But then I also got a chance to meet somebody very cool. So I got a chance to hang out for the day with TS Kelso, who is kind of legendary in the realm of online satellite tracking. He literally was the guy who started tracking satellites online back in the early days. So you're going to hear all about that in a second. But I was with TS Kelso and Josh from AGI, and we did a couple of videos in their studio there. So the first one, which you're going to hear the audio of right here on Miko, is me and Josh talking to TS Kelso about how he got started with his website, which is CelesteTrack. Celestrack.com is a really cool

spot if you want to get data on satellites or if you want to visualize satellites. I love showing it to people to kind of show how orbits work. And it's a place that I often go when I'm doing the MECO headlines to see where things have been put in orbit when there's launches that happen. So we're going to talk to TS Kelso all about the history, how he started that, and then we get into a bunch of topics relevant to today, like debris tracking, debris mitigation, how operators have to work together to figure out when conjunctions are going to happen, how to avoid them happening. And then there's a couple other videos that are going to come out this week on the AGI YouTube channel, where TS is going to show us a demo of

how CelesteTrack works, a demo of some of those visualization tools, and then even talk about Starlink and how you can use CelesteTrack and how you can use the software that AGI makes to figure out when you can see those really cool Starlink trains, when SpaceX does a Starlink launch, how you can track that down really quickly and know if you have a chance to see them in orbit. And TS Kelso is even going to talk about how SpaceX is working with them directly to get them that data before launch so that you'll know when there's a sighting that's going to come up. Because you got to see these things pretty quick if you want to see the train of Starlink.

You got to really get on that within hours of the launch if you want to see it in its full glory. So anyway, this is a long winded intro to what is a really cool conversation about a whole bunch of different topics and some really cool history with the man himself, TS Kelso. So with that said, please enjoy this interview. And don't forget to go over to YouTube AGI's channel. I've got the link to that in the show notes. Subscribe there. You'll see me on there the next couple of days this week, talking with TS Kelso and Josh from AGI and talking about a whole bunch of different topics. So please enjoy. Hey, everybody. Welcome back to SpaceCast. I'm Josh. And today we have two very special guests. This is a special episode. This is like Jetsons meet the Flintstones.

Anthony Colangelo here is a podcaster who does a very popular podcast called Main Engine Cutoff. So welcome, Anthony. Thank you very much for having me out here. And the exciting thing is this is actually going to be, we're going to do this both places. So if you're listening to the podcast, you can hear everything that we cover here and we're also going to put it on the AGI YouTube channel on SpaceCast. People can see what I

look like rather than just hearing what I look like. Yes, and we're going to do a couple of these. So you'll get some cool visuals along to go to go along with that as well and next day Anthony is TS Kelso and if you've ever tracked a satellite or used an app that's tracked a satellite or done anything with one of the things where you see when the satellites go overhead like of radium flares whatever you've probably done used stuff that TS has made available to you. So the fruits of his labor have affected almost everybody on Earth, I think.

That's fair to say that has ever done anything with a satellite. TS Kelso has been running Celestrac for how long have you been doing it for? Almost 35 years. Wow. That's awesome. And we were talking earlier that the TS actually stands for tracking satellites, which is very nice of your parents to name you that. Yes, it's very prescient, right?

They locked you in, though. You had to do this for a living because you had no options. So that's kind of great that you've been doing that for this long. So welcome, both guys. Thanks. forward to a lively discussion here. TS and I were just saying, we just got back from our Christmas party and all the things that we did over the weekend at the AGI holiday party, we can't talk about here. We're not going to cover those events. It doesn't make a lot of sense. It's off the record. But Anthony, I turn it over to you because I know you have never met TS before. TS and I have been working together for a number of years. And you had some questions about the guy, the man, the myth, the legend that runs Celestrac. And it's kind of funny because here at AGI, I've known TS for 20 years probably, close to it. And sometimes I was trying to build him up

and say how many people have come to rely on his data in the world, and it's true. And you forget that when you see somebody every day at work. So it's kind of cool that getting a fresh perspective from someone that hasn't seen TS and like, hey, TS, you're kind of a big deal. Let's talk about what you've been doing for the last 30 years. Yeah, absolutely. It's one of those things, it's the inverse where you don't see a person, then you go, oh, wow, there's somebody behind that thing that I look at all the time. So it's really cool to hang out for a little bit.

And I'm curious particularly about how you got started. You were mentioning before that it goes back to the early days of the Internet itself and even had a different name back then. So I'm curious to hear the early days of Celeste Track and what led you to creating what we have today behind us here. Okay. Well, I mean, I first got started in satellite tracking when Skylab was coming down. And so I had this thing that most people didn't know what it was called a personal computer,

actually a TRS-80 Model 1. And I remember the newscaster coming on and saying, hey, if you go outside, this was around Kansas City, go outside at this time, you can see this dot of light going across the sky, that's Skylab. And so I called one of my friends. We went out, and sure enough, we saw it. And I thought, that's what I want to do with my computer. Now, it turned out that was a little bit harder to do back in, you know, so this was, I think, 1978.

And so I ended up, Air Force sent me to get a graduate degree in space operations at the Air Force Institute of Technology. While I was there, I somehow became aware of a program that would let you predict where satellites were going to be. And so I happened to be in Colorado Springs, which was where they were selling the program. It wasn't like you could really mail order stuff back then. And bought a copy of it, went back, and was trying to reverse engineer it and started asking our class leader who had worked in Cheyenne Mountain. He's like, well, what's this code? I mean, what are these numbers?

Where do they come from? And he said, oh, that's SGP4. And I was like, well, what the heck is that, right? And so I started that education process, got one of the government documents, which a lot of people are aware of now, called Space Track Report No. 3, and started doing that. I was like, well, those are what those values are supposed to be. They're not quite right.

I got it kind of tightened up and everything. But the first question that you had to figure out is, so I have a program and I have a computer. Where do I get the data from? And so they had told you when you bought the software, it's like, oh, well, you write to NASA. Now, this isn't send an email. This is write them a letter. Pony Express. Write them a letter, wait a couple weeks for them to come back

and ask for data on satellites. And so I think the first one I asked for was CSAT because it was one of the ones that was up there that was reasonably bright. And so they would mail you a sheet of paper, first-class mail, so they'd have to, like, photocopy it and then stick it in a big – they didn't even fold it up. It was just like a big envelope. And, you know, a couple days to a week later, you'd get this piece of paper.

And then you sit down, type it up, and put it into the program and do something. And so in the intervening time, I finished my degree at AFIT. I got assigned out to the Blue Cube at Sunnyvale, the Air Force Satellite Control Facility. And so I'm in the Silicon Valley now. And I start working with these guys that are doing electronic bulletin boards. So we had no internet. I mean, the internet existed, but it wasn't really available to people. And so I started working with these guys, and they had software running on an operating system called CPM, which most people probably haven't heard of.

And so we set it up. I had the source code. I could play around with it. But it was always like, well, what am I going to do with this? And then one day I came back from working at the Cube all day and had some more mail from NASA on data and sat down. Because every day I would come back and I'd type these things in. And I suddenly got this genius idea, which upon reflection really doesn't sound genius.

But it's like, hey, I sit down and type these things up, and you have to make sure you don't make mistakes in it and all that kind of stuff. And I bet if I could type them up and just put them on the computer and then people could – Now, remember, this wasn't like get on the internet and connect. This was dial your phone into somebody else's phone, and you had a modem hooked up to your computer with a handset in the modem, and people would call your phone and do that, which is why they call them bulletin boards. I heard that you actually speak facts. Can you do that?

I think I lost that skill somewhere along the way. But so I thought, like, if I just typed it up and put it on the computer, people would come in and download it, and then they wouldn't need to type it in themselves. Now, why I thought there was anybody else on the planet that wanted to do that. Did you let the person know at NASA that just had to keep mailing you stuff every three days? Well eventually right. Because that person had to be getting very annoyed at this point. They're like who is this guy that's mailing us all the time? Well they would send that and so I started

this bulletin board and actually it started in Sunnyvale. I don't really it was in 1985, but where I really count Celestrac starting was when I got to Austin, Texas when I was working on my PhD and finally got things set up and rolling. But we went through that, you know, set it up, you know, people would dial in, they'd get the data, how they found the number is like one of the mysteries, right? Like how do you put out? It's like, hey, we have data on satellites. Come get it.

Just have a guy in the corner dancing in one of those sandwich boys. That might have been more effective. They probably found it because the person at NASA was like, oh, I mailed it a couple of weeks ago. Go check this thing out because that guy keeps asking me for all this data. Well, and of course, so you put it out, and then people would come in and say, hey, could you get data for these satellites? I was like, oh, okay. And then, of course, you have to write NASA, wait for them to start sending it. And finally one day I sent them a letter and I said,

hey, could you add these one or two satellites? And I get this letter back saying, no, I'm sorry. You hit the limit of 25 satellites. We can't send you any more data. And I get this letter back saying, no, I'm sorry. You hit the limit of 25 satellites. We can't send you any more data. And I thought, well, that kind of sucks. And so I wrote him back. How many satellites were there at that time? Well, I wrote him back and I said, hey, this is what I'm doing with the data.

And explained the whole concept because you had to explain the whole bulletin board thing to him and everything. I said, you know, it's probably save you from having other people come in and having to pay to send them stuff. And then they just started sending me like everything they had. They'd send me these stacks that were like a half inch thick that all it was was two line element sets on it. I was like, yeah, I'm not typing all that. But that was kind of the genesis.

And I basically did that for about five years, typing things in until we finally got a way to do that electronically. Yeah. And then you've kind of added on to it over time and recently have added this incredible visualization tool. Is this been a thing that you personally have worked on all this time? Have you had a team at any point? Have there been other contributors or is this a Kelso original? Unfortunately, it's pretty much a Kelso original. Are you looking for interns?

I have been looking for interns. There always seems to be things that make that more difficult than it would seem. But, yeah, I went through a couple of incarnations. We actually, I guess we first got on the internet when I went back to AFIT to teach after my PhD. And I got the FTP site set up so that we could have people come in and FTP data. And so that was a fairly early contribution. This would have been like the late 80s. And I remember when I finally left the faculty in 94, which is about the time World Wide Web was starting up, that the computer guy said, you know what, we just realized that

your site has been the biggest draw of anybody coming into AFIT. I'm like, oh, okay. And it was like by a long shot. It's like, okay. So there were a lot of people coming in for FTP. When I left AFIT, I got approached by a magazine that had just stood up called Satellite Times or was just standing up. And they wanted me to do columns and say, hey, we'll provide, the internet is here. You have the World Wide Web. We have a site. You can put your stuff on the site. And that's kind of where we first got on the World Wide Web. And that went, and so at the time, all the way up to this, we were the Celestial BBS or Celestial RCPM, which was like remote CPM system. Very memorable. Right. Yeah, very memorable. Until the point where finally they folded the magazine after about four years. And it was like, well, okay, so go out and get your own site.

I mean, they didn't tell me that, but that was kind of what happened. And, of course, I went to get Celestial.com. They're like, nope, that's not available. So then it was, hmm, I should make something up that somebody hasn't picked. And I picked Celestirac, and it kind of stuck, and there we are today. I would love to hear a little bit about, you know, I'm not a professional. I talk about this stuff, and I look at the visualizations, and whenever I need to explain orbits to my friends that are wondering how things work,

I'll pull up Celestrak and show them a couple of different satellites, talk about how weather satellites orbit and geostationary orbit and all that kind of stuff. But what is the actual, you know, you're getting all this data. Back in the day, it was from NASA. How do you get that data today from all of the different operators, both here in the U.S., but also globally? You know, there's countries that are very closed-lipped about what they're launching and when, and even the U.S. is that way a lot of times. So what are the data sources that you're pulling in and you know you're kind of the aggregator at this point but that has to come from so many different

places that has to be obviously the most resource intensive part of your work but also to somebody from the outside the most mysterious as well all right so the I mean the primary source we get it from the space track so space track was stood up and I think was 94 so. So we went through a couple of intervals where, so Celestrak did this. I spent many years trying to convince NASA they should do it. Finally, about eight years after first starting that effort, they stood up their own bulletin board system and eventually went on the internet and did that until right about, I guess it was like 2002, 2003. And it was before SpaceTrack came up.

And so we kind of filled the gap there. And then SpaceTrack came along and the Air Force, U.S. Air Force, initially had planned not to allow redistribution. But by that point, Celestrack had so many users, not just amateurs going out to look at satellites going over, but we found educational, commercial, even military users accessing Celestrak to do their missions. And eventually, the Air Force was kind enough to give us an option to do a waiver to redistribute the data. And that's where we get most of the data. But we have, and we can go through that later, we can talk about supplemental TLEs that we get other kinds of data from operators and then convert them into two-line

element sets, TLEs, so that people can use them in most of the software that they use. And that's where, as I said earlier, a lot of these programs reach out to CelesteTrack, pull all the satellite data because it's the best data that you've aggregated, like Anthony was saying, and that's why that it's still today you get how many hits on a regular basis. It's in the millions. Yeah, we run on a daily basis, we run somewhere right around 350,000 unique users. And most of them are just coming in to get data. So they know, and a lot of them are probably using smartphone apps where the developers have come to me and said, hey, I'm developing this app. Is it okay if we grab data from Celestra? I was like, sure. And we try to

give them a little bit of guidance and say, you know, we're not updating it every minute of the day. We do that every two hours. Really, most of the data that we get from SpaceTrack is only updated about every eight hours so if you check you know two or three times a day you're probably good it's interesting to me to consider where we're at you know in the history of space where we have a couple of thousand assets up in orbit a couple thousand dead couple thousand alive and we've got magnitudes more than that license at this point to fly over the next decades um and i'm curious how you see that those data sources shifting over time

do you expect that it will go uh stick in the way that it has been with something like the u.s department of defense uh tracking all these things and providing data for it do you expect it to go more in the realm of making sure operators are filing their flight plans for all intents and purposes? Or do you expect, we've seen in recent years, some tracking systems come online that are more commercially minded, like Leo Labs, people like that,

that have some commercial incentive to track everything that's in orbit. Do you see it as a mix of those things? Do you think we're going to stick in this realm that it's kind of dominated by one player at this point? How does that shift over the next 10 years? That's an excellent question because, I mean, we look at all of that that's going on. And one of the things that we did on Celestrak back in about 2004 was we started up something called Socrates. So it's satellite orbital conjunction reports

assessing threatening encounters in space. So you's Satellite Orbital Conjunction Reports Assessing Threatening Encounters in Space. So you can tell I'm awake. I actually remembered the acronym. Which part did you think of first, the name or the acronym? Actually, Sal Alfano came up with the acronym. So credit to him. And the whole challenge there was, at the time I was trying to show that we could do things

like screen for conjunctions for, you know, between satellites. And I remember we had a lot fewer satellites, you know, 15 years ago than we do now, a lot fewer. I think we were at about 10,000. We're actually almost double that. I have that up here behind us if we look. Okay. So here we are. And I brought up when you started or when you started getting the internet, you said in the 80s, just over 5,000 it looks like. Does that sound about right when you were? When we got to 2004, I remember going back to look and it was something like 10,000 objects on orbit. In 2000.

So now we're at like about 18,000 plus, depending on how you count them. We won't get into that. Yeah, we won't get into that. We only have a little. You can see particular spikes in the data of particular events. But the challenge that we found when we were doing Socrates and part of what we were trying to get people to understand was that there's debris, which you may be able to model reasonably well with things like SGP-4.

But the real challenge is for the U.S. Air Force, for example, the 18th Space Control Squadron, is to track the satellites that maneuver, the ones that are operational and have to maneuver, like if they're in geostationary, to stay in their slot, or to maintain some consistent orbit ground track for remote sensing or something like that. And the challenge that you find there is that you can't do that by passively tracking satellites. So whether it's the Air Force's Space Surveillance Network, the SSN, or LEO Labs, surveillance network, the SSN, or Leo labs. I mean, they can track the bulk of things, but these big satellites that are moving around all the time, we have to have some way to collaborate with the operators. And so where that becomes, you know, interesting today is with operators like SpaceX, who's putting up the Starlink constellation. They've already got 120 satellites. They're talking about doing another launch before the end of the year and launching many of these, you know, in the next year, in 2020.

And now you've got the situation where you have low thrust and the tools that are in place in most of those locations where they don't know what the maneuvers are going to be. There's no way to really accurately predict them. Certainly not for things like conjunction assessment without help from operators. And so we're actually working, we've been working with SpaceX. They provide us their ephemeris. We use that ephemeris to generate what we call supplemental TLEs that are fit using SGP4 to their data so that people can easily ingest them into their software. It gives you, like we're looking at on the SpaceX data, that the fits are normally within a couple hundred meters, like 200 to 300 meters, which is reasonably good.

We'll show everybody how to do that in the video in a few minutes here so we can walk them through the steps. So if they wanted to do that, they can follow along and do that as well. Yeah, and with that whole conjunction analysis, Starlink itself, SpaceX has talked up this ability to load that data into their network and be able to autonomously avoid conjunctions and things like that. and things like that. And that's another side of things that I see shifting in a couple of years is, you know, right now we have people like yourself that are able to alert people and say, you have a conjunction coming up.

One of you two is going to have to move or you're going to have to negotiate that yourselves. That doesn't seem particularly sustainable to me in terms of the growth of traffic. You don't have enough time in a day already to do what you're doing here. And I don't think we can expect you to single-handedly be the superhero that's in charge of, you know, running. We've got a cloning program in place. Is that what it is? We've got nine more

Kelsos growing in the basement. Oh, that's a scary thought. Yeah. So I'm very interested to see what that shakes out to be, you know, and a lot of times people will talk about it in terms of airspace, but that's not really a good analogy because, you know, in the U.S., it's unlikely that we will have someone from Kazakhstan fly over into our airspace very quickly, but you could have a launch out of Baikonur and instantly be impacting constellations or satellites of foreign governments that we don't necessarily have great relationships with. How does that play out? Is there always going to be this centralized kind of reporting system that will alert people of conjunctions? Is it going to go more in the way of autonomous with SpaceX or reasonably probably somewhere in the middle of those two things would be the way it pans out?

But which of those do you see as viable routes and which do you think are kind of a fool's errand at this point? Well, the challenge that I've seen, now, remember, I've been operating what we call the Space Data Center, which is kind of an offshoot of Socrates, now a legal entity that supports 30 operators, I mean, we still have to do things to do quality control because every system has issues, right? You're going along, you're generating ephemeris, you think everything's great, and you miss something in the system. I mean, a file gets corrupted or whatever, and now your data is off for whatever reason. And we have to do this constant QC. And to me, that speaks to you have to have some way to do this in a centralized fashion. I mean, you expect individual operators to do their own QC, and they do. But things still get through. And it doesn't make any – it's not just the operators.

We see the same things out of the U.S. Air Force. We would expect we would see the same things out of any organization. And so the challenge comes to bringing it together. And the other challenge is that while we have those 30 operators talking to each other and sharing data, that we go through this constant operational process to make sure end-to-end the data's coming in, we're checking it, that we do all these other things before we actually do the screening. But there are a lot of operators out there that still aren't talking among themselves. And we've had instances, you probably have seen some of the stories about the Luke

satellite, you know, going around and doing some kind of rendezvous or close proximity operations with other operators. And what we found is that they, I mean, they're not communicating and we can't find out who to talk to, to say, okay, what are you doing? Want to make sure that you don't misinterpret something we have, or we don't misinterpret something you're doing. And we can't get that. In fact, we made some communications with, let's say, third parties involved over there and found out that their response was, well, the picture that you're looking at isn't the picture that we're looking at, which is not really a good response.

Yeah, it's not great. It's like, I get it that you're using a different set of data than we are. We don't have any control. You're not sharing our data, or you're not sharing your data with us. In some cases, we're not sharing our data with them, because some of it is proprietary U.S. Air Force data, right? I should point that out for Anthony, because I'm not sure if you knew it, but the nuanced difference between the SDA and what Kelsey was putting on Celeste track is the data on the

SDA is actually the owner operator data they're sharing together. And that's not going out publicly. So TS takes that and does some analysis on it and alerts them, but it's not a public alert. It goes to the operators and tells them those things too. That's in addition to what's happening on Celeste Track. Celeste Track stuff is mostly from the Air Force but the stuff on the SDA is actually from the real owner operators who in theory have the best data on the knowledge of their

own satellites. Of course, sometimes there is problems knowing after a launch exactly where your satellite is even more so when there's a ride share. We've heard instances in the last year where there's been 60 on a launch and some you never hear from because you don't know which you're supposed to be looking at. When they're that close together, they're also hard to track. Interesting thing is, we could probably do a whole episode on this alone, is even owner-operator data,

that's good enough to do communications with your satellites at geo. If you have a bias in your system, you might be several kilometers off from where you think you are. And for operations, totally fine. Don't even notice. But for CA, that could be a big deal. You might be going in the wrong direction to avoid a close approach

and actually making it worse for yourself. So you have to not only have, like Tiaz was saying, somebody to sort of watch over everybody else, is you have to have even other sensors out there watching to making sure that you don't have biases in your system that are causing you to be good enough for one application but completely wrong – not wrong but within a danger limit for close approach. Right. And your comments about these large deployments is really a challenge, right? So we have, you know, people think that, well, the U.S. Space Surveillance Network has lots of capabilities. It can certainly track things.

You know, the commonly accepted number is down to 10 centimeters, which is the size of a CubeSat, which is why they were designed that size was the thought was, we don't want to make them too small. They can't be tracked. And we can get into a whole different subject there. But when you deploy, say, 100 CubeSats off of a launch and they all look the same to the radar, then this whole issue of which dots on this pass over this radar connect with which dots on this pass over a different radar. And it is a very challenging problem. It's not something that, you know, the U.S. system was designed to do many years ago and is a challenge.

And it's a challenge even for the SpaceX satellites where they're larger, but they're still the same form factor, right? And, oh, by the way, unlike, say, we had like the one Indian launch, we had 88 Planet satellites on it, so three U-cube sats. We worked very closely with Planet to be able to do the association, but they were not maneuvering. And so now when you put Starlink up and they start this low-thrust maneuvering where,'re doing this propagation and you don't know when the maneuvers are occurring and everything, it becomes a very difficult process of associating those dots

so you can get good orbits and then actually doing the identification and everything. So there's plenty to keep us busy for quite a while. But I'm firmly convinced it has to be a team effort, and that kind of suggests that there has to be some kind of centralized process. You can't go off and just say, I know where my satellites are, and then somehow magically assume away the problem of trying to figure out where everybody else's are. Let's not forget the fact that, you know, depending on the size of the catalog that you say now, it's 15,000 to 25,000, depending on who's counting.

By some estimates, that's only 4% of what's up there. And you said the 10 centimeters. If you get lower than that, they've done studies that do sky samples with some shorter wavelength radars, right? Isn't that what they use for those? And they can see a lot more stuff. And so if we have a 25,000, 20,000 object catalog

and we're only seeing 4%, there's, even without new launches, there's a lot more stuff that we should have been worried about the whole time. We just were lucky that we, maybe they are hitting, maybe they're collisions that we don't know about, but there's a lot of stuff up there. And when the new space fence goes online, that's another thing that's going to have to be considered in the mix. And we're trying, we're actually trying to get to the point where, and we're working on this both in the SDC and probably are going to bring some of the prototype stuff

we've been doing in the SDC back on to CelestraX so people can get a better idea of what's available and why some of these issues are important. But we're starting to see that, you know, if you were to do something like bring the space fence online and share all that data, that a lot of people would probably get freaked out to the point where they just did nothing. It's like, oh my God, what am I going to do because there's too many conjunctions? And at that point, we really need to move away from things like missed distance or even probability of collision and start looking at well what's the risk of a collision. If I have a probability of collision let's say one in a thousand

and you say well I should move and it's like well are you moving for something the size of a golf ball or are you moving for something the size of a school bus because I think you might want to prioritize on the school bus and that's and that's where we're going to have to get it to is where you can help people manage what's going on to the point of and there's and there's two reasons for doing this to but to the point of focusing on the important stuff rather than all these other things which today and I don't want to like suggest that I don't think there's a problem, but today we really, even though there are hundreds of thousands of objects we can't see, they don't seem to be causing a significant problem. And so we don't want to lose focus

there, but you've got to be able to get people to realize that at some point, and so we talked about different data causing people to make different decisions. The fact is you can have the same data and still make different decisions, right? Because you have risk-averse people, you have risk-prone people. Tolerant, yeah. Risk-tolerant. Tolerant's nicer, but prone is fine. And so you have people that will look at the same data and make a different decision.

And at some point as a community, as the space community, we need to step up and say, it's okay if you want to let your satellite get hit by a golf ball and maybe lose it because that's just your satellite. But when it starts getting into the, this could be an event that would dramatically impact the entire space environment, then we as a community need to set standards that say, no, if it's going to generate more than this amount of debris or something like that, that there's a cutoff, that you have to do something with your system to try to mitigate that type of occurrence. Yeah, just a good time for a quick plug of Daniel Tragi's new organization called the Space Safety Coalition, spacesafety.org. And so if you have a direct and material interest in satellite preservation of space and not having satellites collide, you can join this organization.

And what they're doing is, I think you guys both know about this, I'm just repeating it for the audiences, they're coming up with rules that the international community uh has agreed upon things about how long satellites should last and and disposing of them and things like that and with all these new problems compounding is this organization is trying to come up with new better rules to adapt to way more satellites and having them come in earlier and just get everybody on the same page with what some of the regulations should be. So check that out. You said SDC and SDA. I don't know if the audience knows the difference. Do we want to explain that? Yeah, we can do that. So I mentioned the SDC, the Space Data Center. The SDC is operated by AGI for the SDA,

is operated by AGI for the SDA, the Space Data Association, which is a group, a legal entity with 30 plus operators involved in it right now. The major operators you can imagine, right? So Planet, Inalsat, SES, Utilsat. I mean, it's an international group. And those 800 satellites, including NASA and NOAA, for example, are all screened every day by the SDC, the Space Data Center. What's the process for getting more involved in that? Is that something that operators are coming to you and saying that they want to be a part of that? Or are you reaching out to other operators? In some cases, they come to us. In other cases, I think in the majority of cases, we're really trying to get people to understand the value proposition. And so it's something that the SDA has worked pretty hard to try to get the message across.

But we can certainly use more help. Yeah. Is there, you know, every time you hear a new story of a new constellation applying for Spectrum or anything like that, Is there, you know, every time you hear a new story of a new constellation applying for Spectrum or anything like that, is that going on lot of lessons learned, I think, along the way that it's important to share. And I think it would be important for some of these large launch providers to step up and actually provide some leadership and a different perspective on things. Because a lot of the operators in the SDA right now are geo, but it turns out we only have, I'll get the numbers wrong, but it's around 280 geo out of that almost 800 satellite. So there's 500 LEOs and, you know, where 240 of them are planet. We have something like 70 from

SPIRE. And so we need to get more of that in, particularly with these larger constellations. You'll bring their perspective in, help each other understand the issues and understand how they expect to work and understand whether that's going to cause any problems. All right, great. Well, thank you guys so much for being here. TS flew in all the way from Hawaii just for this exact moment. There's no other reason. And the party.

No, it was just that. I mean, that was a benefit. They planned the party around his flight. That's right. Exactly. Anthony, thank you so much for being here. Absolutely. Thanks for – it's cool to come out not too far from home and sit with some other space nerds.

You don't get many of that in the Philly area. Yeah, no. We love to have you back and we're going to do a couple more videos here at the same time. So we appreciate you watching and check us out on Main Engine Cutoff, MECO. The address is mainenginecutoff.com

and stay tuned. We're going to talk about some more stuff. We're going to go into some demos and that's it. Anything else, Dad? Yeah, and just the demos that we're going to go into, everyone that's listening to the podcast, check out the show notes over at mainenginecutoff.com and all of the links to the videos

as well in there, or I'll link to them on the blog or something, depending on the release schedule. Awesome. Thanks, everybody. All right. So that was our conversation with TS Kelso about history and current status of Celeste Track and things in the industry. And like I mentioned up front, there's a couple other videos that we shot the same day that'll be coming out on AGI's YouTube channel. So head over there, subscribe to the channel if you want to see the rest of this kind of crossover collaboration. There's some really cool demos of Celeste Track. So if you like what you're hearing about it and you want to learn how to use it from the man himself who makes it, make sure you keep your eyes peeled

on the AGI channel to get a sense of what you can do with Celeste Track. And then, like I said, there's going to be some really cool stuff in there about Starlink and some of the upcoming launches. There's one coming up at the end of the month. So if you want to see those Starlink trains like I very much do, well, number one, you need to not have a cloudy sky, which I unfortunately have had the last couple of launches. But two, you're going to need to know when they're coming overhead with an ability to see them. So that one of the demos is going to show you how to do that. And it's going to be awesome. So I'm excited for that. And I'm excited to have done this whole collaboration with AGI. Like I said in the interview, it's tough to find the space nerds in the Philly area.

So when you can, it's a real treat. And I really enjoyed my day out at AGI. So I hope I can keep in touch with them and do some more stuff in the future because it's a great group of people. And it's cool to have some people in the local area. So anyway, all of that said, thank you all so much to those of you out there who are supporting Main Engine Cutoff. It makes it possible for me to take a day, drive out to AGI and do some of this crossover collaboration work. And I could not do it without your support. So if you like this kind of stuff and you want to see more of it, head over to managingcutoff.com slash support like 352 others have done. Thank you all so much for supporting every single month. This episode was produced by 38 executive producers. Chris, Pat, Matt, George, Brad, Ryan, Nadeem, Peter, Donald, Lee, Chris, Warren, Bob,

Russell, John, Moritz, Joel, Jan, David, Grant, Mike, David, Mintz, Eunice, Rob, Tim Dodd, TheEverDashNut, Frank, Julian, and Lars from Agile Space, Tommy, Adam, Sam, and six anonymous executive producers. Thank you all so much for making this possible. Once again, I could not do it without your support, and I could not do these really cool kinds of things without it. So it makes a huge difference. This is entirely listener-supported, so please head over there if you'd like to help out. And for now, that's all I've got for you. I hope you enjoyed that interview. Keep your eyes peeled for the videos. You can see what I look like, which will be a surprise to some of you. And for now, thank you all so much for listening. I will talk to you next time. Bye.