SIO368: The Euclid Space Telescope Just Launched! What Will It Teach Us and How?

Transcript

This transcript was generated automatically and may contain errors — the audio is the record.

0:00:15 Thomas: Hello and welcome to Serious Inquiries Only. This is episode 368. I'm Thomas Smith and we're back with some astrophysics for you. Ah, love this stuff. Guest expert is Dr. Brian Gillis, back for, I don't know, fourth or fifth time. I'll note this is one of the last of the like stored up ones I have.

So I would love to pretend that this just happened, but you'll see once we start talking about when the Euclid telescope, which is the main subject of today's episode, is going to be launched into space. Talking a few months ago. As of release, July, Telescope has been launched. Brian's going to tell us all about it in past episodes that I hope you've heard. He's talked about his kind of area, his specialty, gravitational lensing, super interesting stuff.

That was all a of the theoretical side, and this is more the practical side. How does the telescope, how does this space telescope actually capture the data that's going to reveal so many cool things about the universe, dark matter, all that stuff. Of interesting. We didn't get into it, but Brian mentions that the Euclid Space Telescope is launching with SpaceX and go into that a little bit, so I won't spoil that.

But one thing I did see that we didn't go into is that is largely a result of the Russian attack on Ukraine. This is a project of the European Space Agency, the telescope was something that was going to go up just of some kind. And then this happened. And so I guess it is fortunate that the launch was compatible with SpaceX Falcon 9 rockets.

So as much as I hate the privatization of space and would much prefer NASA to be at its former glory and for stuff like this to be done by cooperative among different governments and countries and stuff, I guess in this case, it was nice that SpaceX was able to do this rocket. And it was launched July 1st. It didn't explode, but it also has an arrived at its destination, which is a Lagrange point that we'll talk about.

That's still a few months, I think from now, the launch is launched. I another funny note, again, since this took place a little while ago, this is before Where There's Woke, and this is for any research I had done on the James Webb telescope. Is before that daily episode even came out. So if you've listened to Where There's Woke, which I highly encourage you to do, got episode, this recording with Dr. Brian Gillis took place before that.

Coincidentally, we actually talked a decent amount about James Webb, and we were mainly talking about space stuff and not political, social justice -y stuff, which is more about what the James Webb naming controversy is about. More note, get an extra long version of this. Went on a bit of a tangent in the middle about, I don't know, just the most interesting crap in the world to me, which is, is it finite? Does that work? How could there be a big bang if it's that kind of stuff. That kind

0:03:04 Thomas: The middle. Patrons get the full version. Normies, hey, go to patreon.com/seriouspod right now, actually, so you can start that and seamlessly hear it. It's not at the end. Sorry, it happened in the middle. It's a good, I don't know, 15, 20 minutes of really fast. If you want to know all the secrets of the universe, unfortunately, they're behind a paywall. Sorry for that.

That, but hey, unfortunately, um, arguments has been put behind a paywall for me personally since it was stolen from by Andrew Torres and legal bills are more money than I could have even fathomed, so I'm sorry about the monetization, but it also was a bit of a tangent, so it actually makes sense just from a a basic podcast editing perspective. patreon.com/seriouspod if you wanna hear, you know, the answers to the universe and everything. No worries if not.

So much to patrons for supporting the show. Well, I suppose let's talk about the Euclid Space Telescope, how that differs from other telescopes, the science works, how the data is collected, all that stuff. And Dr. Brian Gillis not is just a spectator, uh, has a hand in this, so really cool conversation. As a teaser, there's more Brian coming up. We just recorded recently and was so interesting. Oh, that's the other thing I was going to tell you, folks.

I'm not sure if it'll be this month, but Brian did propose that we do you a bonus content and Armageddon, and I instantly thought that would be the best thing in the world. So I know if it'll be this month, possibly next month. We may get a Science Leoffel Movies of Armageddon with an actual astrophysicist. And so much bad to go around in that movie, but it's so bad good. So looking forward to it.

That is to say, more recordings with Brian coming up, but Euclid Space Telescope. Euclide, of course, was famously an administrator at NASA. No, that's Where There's Woke. I forget that. All right, here we go.

0:04:52 Thomas: Brian Gillis, how are you doing?

0:04:54 Bryan: Doing well today, Thomas. How are you? Oh, certainly out there, it's so today we're following up from the previous episode, so in that we talked about what science we want to do with this space, so this episode is going on to how we get the information to actually do that science. So I thought I'd go through the practicalities of of how the Euclid telescope works, how it observes, and then how we ultimately process all of the data from it. Let's of all, why are we putting a telescope in space? That's the big question. Of so all, it's just cool to say we have a telescope in space.

0:04:56 Thomas: Well. How's space? How's space doing? Wait, who's the dad here? I got to be making these dad jokes. Talk about today?

0:05:37 Thomas: Sure.

0:05:37 Bryan: Light pollution, it is one factor of that, you live in a or near a city, all of the light from that gets up in the atmosphere, bounces back and makes any observations even of the night sky much difficult, as you're just trying to fight with this background, um, there are a lot of ways you can get around that. On Earth, you can go to a pretty remote place and

0:05:51 Thomas: Yeah. Right. Could have a zombie apocalypse. That'll do it. I I've think that was in one of those zombie shows or maybe it was The Last of Us. But just shots the at night of like, oh, yeah, there's a bazillion stars out there when you don't have, you know, humans. So that's one way we could do. We want to do that. Do we want to be a plant zombie apocalypse or.

0:05:59 Bryan: Yeah, that's oh, that's an option, or or we could just, uh, out the middleman and sabotage the electrical grid ourselves. Yeah,

0:06:17 Thomas: Yeah, but if you solve the light pollution, there's this pesky atmosphere, right? Doesn't that cause a bunch of bunch of issues all

0:06:22 Bryan: Uh, the atmosphere is another part of that. So before talking about why into space, let's talk about what we can do on the ground. So best case scenario on the ground, you want to go somewhere as far away from lights as possible, as high up as possible to avoid as much of the atmosphere as you can, and one where the weather is as stable as possible so that you don't have the air above you moving much. And still be a bit because of the atmosphere you do see, but it will reduce it as much as possible. So this is why most of the time when you hear about telescopes, they'll be up on some mountain or another. It's Mauna Kea in Hawaii, which has the benefit that it's out in basically the middle of nowhere in the Pacific.

0:07:09 Thomas: That sounds great, right?

0:07:11 Bryan: Nearby of any note is Honolulu, which is still a fair ways away from Mauna Kea, and the weather there is very calm and reliable, and MaunaKea is a huge mountain. Fact, by some measurements, it's actually the biggest mountain on Earth, if you measure from base of the mountain, thanks to the fact that its base is underwater. Water.

0:07:32 Thomas: Got a dumb question possibly, I've, terms of light pollution, would, if you're on a mountain, but you're right next to a city, but you're way above the city, are you still having problems with light pollution? How does that work?

0:07:44 Bryan: Have some problems. Can be scattered off of many levels in the atmosphere, so if you're above it, a lot of it will be scattered and bounce back before it gets as high as you are, and there'll be less of it that gets up as high as You are on the mountain. So it will still affect you, just the higher up You are, the less it'll affect you. Yeah,

0:08:02 Thomas: But you'd rather, let's say you've got, you an X distance away that you could be, you'd Rather just laterally move out of the way than like go up or down.

0:08:11 Bryan: It's typically better to move away sideways. But even so, the atmosphere itself presents a limiting factor. So one concept I'm going to bring a bunch here is what we call the PSF, which is one of the times astrophysicists resisted the urge to make a really contrived acronym. It stands for the point spread function, which is if you imagine there's just a single point source of light shining into the telescope, how does it actually appear on the image? And when it's coming in through the atmosphere, this is going to blur it out. Basically, there are going to be multiple different paths it can take, which might be scattered and end up hitting the image. So on the ground, it's just going to look like a blur with

0:08:13 Thomas: Gotcha. I never thought about it before.

0:08:55 Thomas: Is the atmosphere where the twinkling comes from? The twinkle

0:08:58 Bryan: Exactly. And in particular, the twinkling comes from the fact that the atmosphere isn't stable. So A like a city, particularly one near hills but not in them, is going to see a very unstable atmosphere above. And so as different temperatures of air drift in and out, it's going to change the PSF, which changes how the stars appear.

0:09:04 Thomas: -hmm Right.

0:09:21 Bryan: And this steady change causes twinkling. Going a bit into the science of this, the reason it changes is that light travels at a slightly different speed through air depending on the temperature. Of The denser air is, the more it slows down light. So when you have different weather systems, one hot, one cold, it'll travel faster through the hot system, slower through the cold, and it'll redirect the light around a bit and causes the twinkling effect we know. So we want to go to a place with a stable weather system, but even there we can't completely get away from it. There's always going to be even steady air is going to blur out the light. And when we want to look at things that are extremely distant away from us, even smallest amount of blurring makes it very hard to do,

0:09:22 Thomas: Mm. -hm right. Right. Right. If

0:10:09 Bryan: Yeah, particularly the type of science we want, which is, as we discussed in the previous episode, we want to measure the shapes of galaxies in order to calculate how much they're being lensed by gravity. And so we don't want what's essentially just lensing by the atmosphere to get in the way of that as we're just adding one extra unknown on top of it.

0:10:29 Thomas: You could build a telescope on top of Mount Everest. Would that be pretty, pretty sweet? Would that be even?

0:10:35 Bryan: Think the problem is that Everest doesn't have a very stable weather system there.

0:10:40 Thomas: All know that I've seen it referenced that, like, when know how much this is exaggerated in photos or whatever, but I've seen photos posted of like you get to the top and it almost looks like you're in space because there's so much less atmosphere, but That might be an exaggeration.

0:10:51 Bryan: Yeah. That's certainly going to be an effect. But I think the stability you get around MaunaKea outweighs Everest. And there are also a lot of other factors like political considerations and that Mauna Kea is in the U.S.

0:10:52 Thomas: Right.

0:11:05 Thomas: And If Mont Everest is a republican, you don't know, maybe a good point. I didn't even think about that. Oh,

0:11:08 Bryan: Yeah. Yeah. So actually, the problem with Mauna kea is that it's considered sacred to many Of the Native American tribes in Hawaii. So they of tolerated or disliked the building of a lot of telescopes that were put there and have kind of gotten fed up recently and are completely blocking any new ones. And this one of those situations where there's really no right answer. How Do you balance a culture feeling that something is sacred to them and should remain untouched versus the desires of science to learn more about the universe? It's just uncomparables. And so so existing telescopes there have continued operations, but other things have ground to a halt. So decided to look at other places. There's been a lot of telescopes built in Chile, which also has a friendly government and has good mountains for this.

0:11:14 Thomas: Yeah. Hmm. Right.

0:12:02 Thomas: About that one in, I think it was at Puerto Rico, the one that like got, was in Goldeneye, the movie, but then, uh, fell into disrepair. You know which one I'm talking about?

0:12:09 Bryan: Oh, no, I'm not familiar with That one, actually. I'd have to look it up. Yeah.

0:12:13 Thomas: Oh, Sometime around, uh. I think as a result of the... Maybe that hurricane. Um. There's footage of it just totally collapsing.

0:12:20 Bryan: Yeah, I'm just doing a search here, the one that shows up is Arecibo Observatory, yeah, is, uh, think a radio telescope.

0:12:27 Thomas: Yeah. Yeah. Radio telescope. Okay. A little different, but, uh uh, so

0:12:27 Bryan: Yeah. Yeah. Yeah, I don't work with radio, so it didn't Come to my mind, but yeah. That one's there.

0:12:33 Thomas: You don't even care that it just died, huh? I thought it was a pretty traumatic event. I, I, I'll admit I shed a tear because it was just like this big scientific endeavor.

0:12:40 Bryan: Now The thing About astrophysicists Is we're all basically racists when it comes to wavelengths. The the visible spectrum people could care less if the infrared people die Or The radio. Oh, that's even more out there.

0:12:51 Thomas: When it comes to wavelength. Uh, that's funny, but that Joe, very funny joking aside. It, I found that really sad. I'm, I'm I, uh, it was just one of those instances of like, yeah, this thing creation of science is now just completely collapsing. And I don't remember exactly how it happened. I think I read a thing about it, but I found it depressing. Okay. So not that you're, you're racist against that one. Well, spec spectrum mist. I don'T know what, yeah.

0:12:59 Bryan: Yeah. Oh, yeah. Yeah. Yeah.

0:13:17 Thomas: Wave, wave, wavelength this.

0:13:18 Bryan: Spectrum. There we go. Going through all this, the specific type of science we want to do, almost any atmospheric interference makes it very difficult. Occult. So we decided that this is one of the cases where all the effort into putting something into space is worth it. Once you get above the atmosphere, the PSF gets reduced dramatically. Now, it doesn't go down to zero because the atmosphere isn't the only factor that causes causes a spreading Or blurring of light. We face is that light is a wave and waves like to spread out. Even when you think you have light going in a straight line, which when you have stars halfway across the observable universe and we're getting light from them, it's all practically in a straight line, as soon as it hits some barrier and goes past it, the wave will spread out on either side of this.

So you can think it's kind of like if you have a dock by a bay and that's going out perpendicular to the waves. Once the waves come past it, it'll block off part of it, and then by the side of it the wave will go around, but as soon soon as it hits that corner, it'll spread out again from that corner. And light does the same thing. Since it goes into a finite area in the opening of a telescope, it wants to spread out from there.

So there's always going to be some notable PSF. And our goal is to try to make it as stable as possible and as small as possible so we can work with it. And a lot of what we did in Euclid and how it was designed was built around making sure that the PSF will be basically unchanging.

We do things like, instead of planning to put Euclide into orbit directly around the Earth, we put it into one of these what we call Lagrange points, which are these stable positions in orbit.

0:13:19 Thomas: Yeah.

0:13:46 Thomas: Love.

0:15:20 Bryan: L2 has always been my favorite.

0:15:23 Thomas: Yeah. L2. Love L2.

0:15:24 Bryan: No.

0:15:25 Thomas: L2. Is good. I go for L3? I don't remember which there. I do remember learning about these, and there's some cool these where there's sort of like an equilibrium point or something like that based on the different competing kind of forces of gravity. Hmm.

0:15:25 Bryan: Yeah.

0:15:39 Bryan: So basically how it works is, let's say you've got one big massive object at the center of a system and then one less massive but still notable object. So let's, like, we've got the the Earth and the Moon. The Earth dominates, the Moon is less so, and mathematically it has to be less than I think about 13% as massive for all Lagrange points to. And then the Moon is orbiting around the Earth, and then it turns out there's a total of five points in this system that move around, relatively speaking, where some amount of stability where the force of the Earth and the Moon's gravity will cancel out, and something that's at or in a particular type of orbit around one of these points will stay there, again, relatively speaking.

0:16:27 Thomas: Right.

0:16:28 Bryan: So there's one, the easiest thing is just between the Earth and the moon.

0:16:30 Thomas: Between Between them. Yeah. It straightforwardly mathematically the ratio of their masses kind of thing? Like it's going to Be, you know, if the moon is one sixth, the earth, is a gonna be, you five sixths over to the moon basically in that for that Lagrange point or is it more complicated than that?

0:16:46 Bryan: Gravity goes off as the square of the distance, so it doesn't vary quite linearly. I think you have to account for that, and I believe there

0:16:49 Thomas: Well, Doesn't That cancel Though, I mean, because it's they're both

0:16:55 Bryan: Um, I would have to double check the math, but there are also some complications to the fact that we're working in a rotating frame of reference. Which means that you have to add these fictional forces, so

0:17:02 Thomas: Sure. Yeah. Am sure it is massively more complicated than that, but like, roughly speaking, it is like somewhere where the two forces are roughly offsetting, and Then you get, it is a nice comfortable zone. I go there all the time. I vacation, lounge Around there is no gravity, it is great.

0:17:08 Bryan: Yeah. Yeah, yeah, Yeah, Land around is actually the key aspect to it, as most of these points are actually unstable equilibrium, and that if you were to put something there, if you're offset by even like one Plank length from it, it'll eventually drift out of that position and keep going farther and farther away. But it turns out there are actually some set of complicated orbits around these positions that are stable, so that's what we're actually trying to get the telescope into, is a one of these orbits around that Lagrange point. And it also has the benefit that you can have more than one thing there, we don't just have one taking up space, and then anybody else who wants to use it has to wait in line, yeah.

0:17:34 Thomas: Really? Really? Yeah. There's Exactly one telescope shaped Lagrange point, and you put it exactly in there, like a video game, like it fits in perfect, and then nothing else can go. Well, surprised that it's, it's sensitive, I, things even that precise when that just a hypothetical, you know, if you're talking about the plank length length, there's the some hypothetical point where it's like perfectly stable, but that sounds like you would wouldn't really achieve that in real life.

0:18:18 Bryan: Yeah. No. It's Going to go off, because, um, like, to joke that we assume everything is perfectly spherical, but the earth isn't a perfect sphere, the moon isn't a perfect spear, they're a bit squashed, first of all, um, due to their own rotation, and then you also have the influence of the sun, and the fact that these aren't just orbiting each each other, they are orbiting in a vacuum, literally, but metaphorically, they're not orbiting in the vacuum. Any of all of this, or one of the drifting other plants, is going to cause it to drift off. So yeah, there is no one precise point it will stay.

0:18:23 Thomas: Yeah. Right. Mm. -hmm. Right.

0:18:57 Thomas: Right, so, but does the telescope have any like little thrusters or anything to keep it in in line, or is it able to do that some other way, right?

0:19:03 Bryan: It does have thrusters. We try to avoid using those as much as possible as we use up a limited resource. And this is all baked into the planning for the mission, that we've got enough ready for about a-year mission. If things go well, it doesn't end up going off too much. That can be extended. Also, we try to do fine-tuned sensors with gyros, as we call them, which is basically using conservation of angular momentum to forcibly spin one part of it to force the rest of the body to spin in the opposite direction, so that we can do fine, uh, adjustments without shooting anything out from it and losing that. By using gyros, we just use up some energy, um, is less finite, uh,

0:19:11 Thomas: Mm. Hmm, yeah. Now I'm forgetting, what was the timeline on this? Because I remember the James Webb, I want to say that had to be planned for, I don't know, a hundred million years, because the amount of pre-planning you have to do for this stuff, because of all these variables that you're talking about, isn't it just just pretty extreme? Cause if you mess one thing up, you know, ever since they messed up that, that damn mirror on the Hubble, you know, then they're like, well, shit, now we got to go all the way up there. Now I got to, goodbye, I just came from there, I got to go all the way back. So I feel like this stuff takes quite a while to plan out.

0:20:21 Bryan: Usually, any space machine, you're talking about a five to 10 year lead time. And for Euclid, it's been a bit over 10 years, I think, since the initial planning. And a lot of that is in construction. And then there have been multiple delays to it. I think it's been delayed about three years since the planned launch. And those delays are because something wasn't exactly perfectly right. And because, of course, as you say, once you put something up in space, you can't just pop over to it to do fine adjustments. You do a specialized servicing mission.

0:20:48 Thomas: Yeah.

0:20:57 Thomas: Especially in L2. Am I right?

0:20:59 Bryan: Yeah, that's a tough one. But Hubble was actually kind of unique, that you could service it, because it was in Earth orbit, and it was important enough that they were able to do some servicing missions. But something as far out as L2, yeah, not doing any manned missions there. So if something breaks, that's that's it. It's broken. If, if the rest of it still works, you can run the rest of the telescope, but if not, that, that's it, mission over, yeah.

0:21:06 Thomas: Right.

0:21:24 Thomas: I remind me. So who responsible for this? Is this because James Webb was was NASA, right? Right. But who's now

0:21:30 Bryan: So, Uh, is the responsibility of ESA, uh, European, the equivalent to NASA, um,

0:21:35 Thomas: Do they have advantages? Because I know a lot of the tale of James Webb was trying to convince Republicans that we should spend, you know, 50 cents on the most mind blowing, amazing science we could possibly do while they want to put, you know, $50 trillion into a of war that, you either we won't use or we'll kill a bunch of people. Now, do we do they have that same problem over at ESA or is it a little easier to get stuff done?

0:21:58 Bryan: Yeah. Not, not really, um, the the US is where like most of the world's military spending is. So, yeah, due to all of those geopolitical concerns, Europe doesn't spend nearly as much on military and is as much more money left around to put into science, and can get things like this done with, with overall a lot less hassle. And especially since it's done by multiple countries getting together, a change in one country's government can't really scupper the whole project.

0:22:27 Thomas: Mm-hmm. Hmm. Hmm.

0:22:33 Bryan: And most, they'll cut out funding to one aspect of it. But the rest of the countries, if it's important enough, they'll figure out a way to keep it going nevertheless. Actually On that note, a bit about the politics of it. The way they decided to the mission is that each country would be handling one particular aspect of either the design, building of the instrument, or the science processing. So, for instance, the UK is primarily in charge of the shear measurement, with a bit of help from Germany in that. Um, I believe, uh, France is doing the simulations, um,

0:22:33 Thomas: Right. Right? I got to imagine that somebody's got to be in charge of piecing all that together. And that's it's gotta be, uh, was that, Mars that mission. I remember back in the day where allegedly there was just the difference between metric and you know the, the, whatever the hell the queen's fucking system, what is it? Imperial system caused a, caused a problem there. Are there challenges like that? Or is it pretty easy to get to, uh to work together?

0:23:27 Bryan: Yeah. Yeah. Imperial. Oh. Some challenges, uh, that get into it, but a lot of this, uh, agreed upon, uh, beforehand. Like, yeah, we've all agreed to use the metric system and we've all agreed to speak English in meetings with each other, though that doesn't, um, the, I don't know what it is, but when searching our website, whenever I find meeting notes from the French, they're always in French, so Um, their internal ones, they just insist on doing their internal meetings in French.

0:24:03 Bryan: And all of their notes, but everybody else is speaking English, yeah.

0:23:57 Thomas: Sure. They've got, I know Know, they've got so many rules on wine, you, you know, can even call yourself or be or produce wine in one of their appellations without it being an exact blend or whatever, so maybe they got the same thing with science, they're like, I'm sorry, it's not science unless it comes from the, the Champagne region of France

0:24:20 Bryan: You can't call it the Euclid space Telescope unless it comes from the Euclid region of France, yeah.

0:24:22 Thomas: Yeah. Exactly. Like it has to be in France, I'm sorry, you can convert on your end, but it's non-science unless, uh, it's in French. I wouldn't be surprised.

0:24:28 Bryan: Yeah. All of that goes together. We get the telescope up there. Let's talk about what data we're taking. So the overall, the goal is to take images of about one third of the sky in total, which is about the most that we can get good images of. When you go out really deep, the small amount of interference that comes in from both the solar system and the galactic disk turns out to be big problems when you're looking this deep.

And so once you mask that out, you've got maybe one third of the sky left that you can look at. So they set up the telescope to do what they call a stop and stare pattern, where they'll use the gyros to adjust it a bit and then hold it in place, looking in one particular direction and basically just take photographs of it through the telescope.

To give you an idea of how deep these are, it, to, let's say, the camera on your phone, which has an aperture of maybe one square centimeter or less. The Euclid telescope has a one square meter collecting area.

So with the squares in there, that's a factor of 10,000 more, and the phone, uh, a picture in maybe 1/60th of a second, the Euclid's pictures take about 10 minutes, uh, so, you know, overall we're taking, uh, that are more than a million times deeper than what you would get, uh, a cell phone camera.

0:26:00 Thomas: Can I just take a million photos on my phone and then send them to you guys? Would that, would that work or no?

0:26:00 Bryan: Um, No, if you, if you, it from space, yes. Certainly.

0:26:05 Thomas: No. Okay. Oh, you're right from space. All right. I'll throw my phone really hard.

0:26:07 Bryan: Yeah. Yeah. Now there also are some differences in the technology that, uh, the cameras phones use a different type of, uh, than we have, uh, the telescope, um, what's up in the telescope is called a CCD, which was used for phone cameras for a while, but, um, as practical to build as, uh, they use, but it does have the advantage that it has a lot less noise in the images than the technology that goes into phone cameras. Still does have its quirks that we need to worry about. One interesting problem is that while the telescope is up there for many years, every once in a while, a hyper-energetic photon is going to come hit it and slightly damage the camera's surface.

0:26:48 Thomas: And all know that photon, one photon, that overachieving photon coming from some, you know, galaxy of bazillion years in the past. It's like, I know I'm still doing my job. I know my, where I came from doesn't even exist anymore, but I'm still gonna do my job.

0:26:54 Bryan: Yeah. Fred.

0:27:05 Bryan: Yeah.

0:27:06 Thomas: Take pride in your work.

0:27:06 Bryan: Yeah. Your work is damaging the quality of scientific images, take pride in it. That one photon comes in, damages one pixel of the image. So what happens in these CCDs is we read off each rows. We basically transfer all rows over to the side, read them off, then take that data off. And what happens is one of those pixels gets damaged. So each time we transfer it through, that pixel basically gets a bit greedy. And while the electrons that we've built up, which are basically the signal from the light, one damage pixel is going to be a bit greedy and is going grab onto some electrons that don't belong to it.

And when we try to transfer them over, it's just going to keep on them and they'll get trapped behind. And the net result of of this is that we end up with a streak off that eventually attempts later we finally claw uh electron free but we get it, it's over in the wrong pixel now, so all of this works together that once the uh telescope has been up there for a while, the the raw images look very streaky, like any source will have a streak off to one side of this.

0:27:12 Thomas: Yeah.

0:27:47 Thomas: Is Hmm.

0:28:15 Bryan: And so there are actually some good images of this you can find from how this happened to the Hubble Space Telescope, which uses the same technology and faced the same issue that they noticed once it had been up there for a while, the images were getting streaky and they had to figure out a way to handle it.

0:28:32 Thomas: And this is just one damn photon could cause this?

0:28:35 Bryan: One could cause it and will cause a trap that we cause it, and then from that point on any images taken into the future will have a streak at least along that row, so it it only caused it so

0:28:46 Thomas: What's a row? Oh, just like in terms of like...

0:28:49 Bryan: Yeah we have a grid of pixels with rows and columns and

0:28:49 Thomas: Oh, gotcha. That's gotta gotta be a shitload of pixels. What's the resolution on this bad boy? It 1080p or?

0:28:59 Bryan: Actually a fair bit more than that, so we, we have an array, a 6x6 array of detectors, so 36 detectors. Each of those is about 4,000x4,000 pixels, so and then all told, this is covering one of a square degree of sky. Give you an idea, that's about the size of how the moon appears on the night's sky for a single image. And we're taking a picture of that much of the sky in this 24K by 24K pixel camera.

0:29:34 Thomas: Thought you were doing a third of the sky though. Or and Gotcha. Gotcha.

0:29:35 Bryan: Is Well, in one step. Yeah.

0:29:38 Thomas: So photon does. I didn't totally understand all the science terms there, but doesn't some sort of like lasting damage to where your images are messed up for for what? Forever.

0:29:47 Bryan: For the rest of mission, it'll only affect one row of one detector and only everything to one side of it. So just a bit in the way these CCDs work, that way they read off the images is they basically do it one column at a time. So imagine you've got a grid rid of pixels, you first read off everything in the leftmost column, and then shift all the information one pixel to the left, read off the left most column again, and repeat. So what this one jerk a photon does is it goes and hits somewhere in the middle, and one of these pixels gets a bit sticky, a bit grabby, so that when it's doing this process of shifting over once it gets greedy. It grabs on to the electrons that are supposed to be transferred over and hold on to them until they're eventually jostled free. So this results in that every image you take with that detector is going to have a streak in that row.

0:30:26 Thomas: Hmm.

0:30:43 Thomas: Let it's like reading. Is this in the actual, like, I don't know, lens of the thing? Or is this in reading the data of what it's picking up or something?

0:30:55 Bryan: So this is in the detector. So let's just clarify on that what's going on. Has a lens, first of all, which light comes in, and then it hits some mirrors, which focus it. Then it hits a beam splitter, actually. So one other part of this I've glossed over is that we're actually doing two things with the light at once. One is taking these high -quality images, and another is sending light different wavelengths over to a different device, which tries to figure out how far away these objects are.

So for now, we're just talking about the image side of things. So think about it this way. My light goes into the lens, hits a mirror, goes through a beam splitter, which it filters out so we only have visible wave lengths, then bounces around a few mirrors, and then eventually hits a plate that we call a detector, which is similar technology to what's in your And on this, this has a grid of pixels.

And then when the photons that we want to observe hit that grid of pixels, whichever spot they hit, they'll basically knock up an electron that will be caught on the other side of it. And that electron will persist until the end of the exposure. And then at that point, we go through and read off how many electrons are in each pixel. So do you follow along there?

0:30:57 Thomas: Detector. What's a detector? I think so. Electrons in a pixel. Jeez, how many electrons are in a pencil? How many can you fit in one pixel?

0:32:19 Bryan: Yeah. Well, that depends on how much light hits it. The point of this detector is that you've got some sort of semiconductor agent. Excuse me if I'm not using the best terms. This isn't quite my expertise, but yeah.

0:32:24 Thomas: Ah.

0:32:34 Thomas: How would I know? I don' t know.

0:32:36 Bryan: Yeah. So, so think of it this way. Let's just say you've got a two by two grid of pixels. You've got a square metal plate, and this is set up so that if a ray of light hits that plate, it'll excite one of the electrons from one of the atoms, and that'll flip over to be on the other side of that plate until it eventually gets read off.

0:32:45 Thomas: Okay.

0:32:59 Thomas: Is it like one of the, oh no, I'm not gonna know what this was called. Remember that thing as kids? I, those metal pegs, and it was in a a little square of them, and you shove your face into it, and then the metal pegs on the back like stick up and you know, that is, it kind of like that.

0:32:59 Bryan: Um, Yeah. Yeah. I know what you're talking about. Yeah, that's a good way to think about that, like of putting your face into it. You're, and every time a rock hits, it sticks one of those pegs out.

0:33:20 Thomas: There, you go okay, I like that, but color factor into this at all or, or is it just like, if, if there's a photon, we want to measure that. And so you're, you're getting some idea of like the intensity of that.

0:33:33 Bryan: Answer is yes, but it only factors in the way we've designed it to factor in. So what's going on here? So, in a physics perspective, we just think of color as the wavelength of light. When the light is coming in, the color comes in a few places. So first of all, the lens, you, nature of that material is only going to allow certain wavelengths to pass through. Some are just going to be blocked off. Some are going to be reflected off.

Some are going to be absorbed, and some are going to come through. So now we only care about the wavelengths that come through, then the next is what we call a dichroic filter. This is something we put into deliberately split up different wavelengths of light. So, the visible wavelengths of light are ones that we have very good technology to detect and get clear images on.

So, this dichroic filter takes those wavelengths of lights and reflects them off, and then other wavelengths that can do something with these are what we call the near infrared, these are basically redder than red wavelengths, ones that the human eye can't detect but are out there, and we can design technology to detect those pass through the filter and go on to a different instrument which handles things a bit differently to try to figure out how far away things are.

So basically we sort things according to color, and then once they hit the detector, there's a bit of a different response, like the chance that it'll be detected varies a bit depending on wavelength. The net result is that we get what's essentially a black and white image of all the combined visible wavelengths that we see in that one picture.

0:35:12 Thomas: Are you capturing it in a way, is that, is and it's not, I know it's not you up there personally, is that the telescope capturing it in a that you can like reconstruct the color for us to look at for fun, or is that data like not captured in there and you're not worried about it or something?

0:35:28 Bryan: For this device, it's basically not captured. For the visible device, the infrared device does have ways of capturing color data, but it's not really set up to be nice to look at. It's set up up to, um, first of all, it's not even looking at visible color, so it's not what a human would see looking in disguise. It's just, yeah, different batches.

0:35:47 Thomas: The Infrared infrared one, right? Okay. Cause I was going to say, cause this, this telescope is though, isn't the main, data the that you're getting. I thought you said it was the visible, right?

0:35:51 Bryan: Yeah. So the main data is, um, band, so we just say that's short for, uh, visible light spectrum, but effectively it all gets blended together and all the information about which wavelength is which gets lost. And so, we have that. The infrared does get split up a bit according to wavelength. So, you could, if you wanted, create images that have the infrared color and get what we call false color images, that it isn't how a human would see it, but you can add color information to it, and It will be real data there.

0:36:05 Thomas: Hmm. Okay. Yeah. Yeah, that's, you know, that is always interesting. I, in some ways it's going to sound dumb, but time the I discovered this, it made me kind of sad because like all the pictures, the coolest space pictures you see from whatever telescope, you you know, those big old giant, uh, don't know, nebula looking things or whatever those are. I, weird in one sense, it is an image from space and it is out there and it is, you know stuff that is there on another sense. It's like, but I would never see that because it's using a bunch of non -visible wavelengths of, of light. Right. So it's kind of interesting to think about just as a person, like, how do you, how Do you think about that? Just as somebody who likes cool space stuff.

0:37:03 Bryan: Well, I'm going to tell you a secret, Thomas, in that pretty much any picture that goes through a computer does the same thing. It only has certain bunches of light that get caught in certain color channels and are reconstructed, and now for the ones you see, with like, say, your camera phone for digital images, particularly some old analog images are more true color, but for digital images it gets split up into three color channels, and we do everything possible to make them appear as realistic as possible to humans when it comes out the other end.

0:37:41 Thomas: Okay. Well, no, I wasn't really talking about that. I was talking about, um, fact that it's picking up light that we would never even see because it's like, you know, those wavelengths that our eyes don't even. So like some of those.

0:37:48 Bryan: Yeah. Well, to me, I see that, I see that as even cooler in that we're seeing something we can't see. We're just translating it into the visible spectrum.

0:38:00 Thomas: Yeah. I don't know why in my dumb brain though. I'm always like, yeah, but I wanted to, I wanted TO think that like, when I see a cool picture from a telescope, I wanted To think like, Like if I was out there in a spaceship looking through the, you know, the window that I would see that, but that's not really true. I would see not even close to that.

0:38:16 Bryan: There are some certainly that take pictures of visible light and get close to what you see. Though, unlike pictures that we take on the ground, the filters we choose for visible light aren't selected to match human vision.

0:38:29 Thomas: Interesting.

0:38:29 Bryan: There are various different filter sets, and none of them really match up perfectly with human vision. We use a set of four or five different filters that select different subsets of visible light. Whereas humans, well, at least for color vision, split it up into three, and then there's also the separate detector for black and white vision, which is kind of, so you can think about this, when that you're seeing at night, you're using the, I think the rods are the low light ones.

0:38:58 Thomas: Oh yeah. The rods and cones.

0:38:59 Bryan: Ones, so you're using the low light, uh, perceptors, perceptors in your eyes, which are, you, they collect all the the light and don't get any color image. Um, that's basically what we're doing in Euclid is we've just got the one broadband, uh, so this is like what you would see with your night vision. Um,

0:39:14 Thomas: Oh, okay. That's interesting. Gotcha. But on the other hand, it's also cool when you talk about, you of those famous images. What's the one I'm thinking of? That's like the the brown wispy cool giant thingy, uh, what the hell maybe it's that like isn't it true that you wouldn't see any of that, like if you were just, you know, somehow within view of that, you wouldn't actually see it because all those, there's a lot of like infrared stuff, yeah

0:39:28 Bryan: Horsehead nebula maybe, yeah or true, a lot, lot of what you see in these interesting images is from uh a of light that aren't visible to the human eye, yeah

0:39:47 Thomas: I just think it's interesting, I don't know, for for whatever reason when I first found that out I was like no, because I wanted to feel like there is a big gold brown horse head nebula out there, but but like there is, but on the other hand, there's nothing you know special or sacred about the light that we can see, it just happens to be what we evolved to care about on this planet, you know, because I don't know, predators or prey or whatever, so like on the another hand, it is still all there, like that it's not that it is fake, it's just kind of funny, I, I isn't so much a, like, like a physics thing, is more just a human, you, and also emotional thing, which is like, well, but I wouldn't be able to see that if I was out there. So for some reason, it feels weird to me.

0:39:58 Bryan: Yeah hmm I understand that, and I always just see it as the other way, that uh is illuminating the invisible for us like one example is some creatures on earth do have expanded wave lengths and there are other creatures that take advantage of this, like I've heard that supposedly if you take a picture of a spider's web in ultraviolet, you'll see flower-like patterns on it because the the spiders take advantage of it, that um insects, uh, see into the UV, and so they draw designs on them to attract those insects

0:40:36 Thomas: Right. Right.

0:41:02 Thomas: But on the other hand, it's also interesting that you, I think you need some amount of limitation, right? Because if I Was on Earth and I had eyes that somehow saw in every single wavelength, like Like there's no limit at all. Would that just be like overwhelmingly impossible to make anything out? Like, isn't just, I'd see radio too, right? It was just everything.

0:41:20 Bryan: Well, brain could maybe be set up to handle it, but there are other reasons, like we talked before about those jerk photons that cause damage. Once things get to too high energy, they start to be damaging. And we see it like with ultraviolet, that it's damaging just to our skin to be out of it. And when you try to see it, you have to evolve differently to be able to handle it. And there's also a benefit to being able to separate out different colors and have different detectors. Connectors.

0:41:26 Thomas: Those jerks, right.

0:41:45 Thomas: Yeah. Yeah. So, okay. All this is to say, I know this is weird Thomas question spirals of nothing, but I real actually question here is, I was curious that you said you were even with the telescope filtering out certain wavelengths and doing like a visible, because in of the science you are wanting to do, which we talked about, you know, gravitational lensing and I'm sure a lot of other stuff, I guess I would have thought like, yeah, well, any, wavelength, any data at all would be useful for that, wouldn't it? Or why is it that you end up limiting it in certain ways?

0:41:51 Bryan: Yeah.

0:42:16 Bryan: So this is where the science gets a bit complicated, and I think there are other reasons that go into it, but yeah, it's just

0:42:20 Thomas: Oh, just now, just nowadays where it gets complicated, right? It's been easy up until now.

0:42:25 Bryan: Now it gets a BIT complicated. Okay, so remember we talked before about the PSF in space, that light is a wave, once it goes past something it spreads out. Now, the trick is that the way it spreads OUT depends on the wavelength of the the light. So if you have a wavelength of light, you can very easily model how that wavelength of light will spread out and what PSF you'll get from it. But when you have the broadband filter, it's going to be some mix of light of different wavelengths. And so the way it will spread out depends on what light you got in. Like if the source is actually very blue, it's going look different from if the source is actually very red, but you don't know it in the final image. So the broader the spectrum you look at, the more uncertainty there is in what the underlying image actually Looks like. So you have to do this balancing act between you want to get more data so you have as much as possible, but without making it too much that the uncertainty in the PSF starts to overwhelm things. Yeah.

0:42:36 Thomas: Hmm. Hmm. Huh.

0:43:32 Thomas: I just look up at the stars, I just see a star, right? I mean, I just see or a galaxy or whatever the hell it is. Um, my brain compensating for that? Or is it just because I'm only looking at a certain, you and because this is looking so much further, you or space, that's you need to know how much the stuff has kind of spread out or PC, whatever you said.

0:43:54 Bryan: A yeah. So this is something that is usually, um, small, um, effect that it spreads out over such a small, um, that most of the time the human brain doesn't notice it, and basically the only time you do is looking at, um, stars up in the sky and you see them twinkling, and The twinkling is the changing of the PSV, so Yeah, the human mind doesn't compensate for it, but when it's big enough, you do notice it.

0:43:55 Thomas: Yeah. Oh, okay. Gotcha. But because the telescope is just so much more, I precise or looking so much deeper. So this telescope is looking somewhere where I, if I just looked up and ignore all the, you know atmosphere sphere and the light pollution, if I was in space and I looked there personally, would I just see basically night sky, just dark, but because this thing is so sensitive and amazing, it's going to see just tons and tons of galaxies and stuff. Right.

0:44:45 Bryan: You'll see tons of galaxies, and yeah, this is the part where the fact that this is an audio medium works against it, But there are plenty of images you can see online of deep fields, like the James Webb telescope has taken tons of good ones recently, much any spot that That you see an from like a Deep Field or James Webb would look to the human eye completely empty.

0:44:56 Thomas: Gosh.

0:45:07 Thomas: It's hard to fathom how much crap there is out there. Like

0:45:11 Bryan: Basically taking images with the telescope, we've got to all the point that, uh, we're reading out images. These get transmitted, uh, down, uh, Earth, and then we have a whole team that goes to to work on processing these images. So not quite where I come in, but it's closer to what these are, people I work with. So the first team is basically just doing what we call reduction on the images, that it's a standard cleaning them up.

So one of the things that needs to happen is that different parts of the detectors have different sensitivities, and some will like always show an extra one pixel in this corner, corner, and we subtract that off. We do what's called flat fielding, so we get rid of all the persistent effects on the detector. We look for other defects that can come up, like cosmic rays, which are these energetic jerk photons that come in and disrupt things. Not all of them will damage the telescope.

Some of them will just cause an image to show up in the detector, and in fact, you get a lot of this if you look at, um, images, you'll see a lot of little white streaks on them, and these are all cosmic rays that hit the telescope, uh, it was taking the image. So in order to handle these, what we usually do is we just take multiple pictures of

0:46:28 Bryan: The same spot, yeah.

0:46:28 Thomas: Just Photoshop. Oh, okay.

0:46:30 Bryan: Yeah. We

0:46:30 Thomas: You just hit that on Insta, and you blur, and it's a blur tool, and then it smooths it out. Makes the universe look younger, too. It looks like only 12 trillion years old.

0:46:39 Bryan: We're Taking out all the wrinkles, yeah.

0:46:41 Thomas: The universe is also catfishing, by the way. It's on Tinder, and it's like, I'm only, you know, 11 trillion years old.

0:46:46 Bryan: Only 12 Billion Years old.

0:46:51 Thomas: Oh, billion. I don't know why I said trillion. Yeah, 12 billion years old, and then, yeah, getting into a lot of trouble when, you You know, you actually see the real photos instead of the telescopic ones you've you've made up. So you're this something that you're estimating or is it something you can actually filter out? Like, here's what the real image should be somehow.

0:47:01 Bryan: Yeah.

0:47:10 Bryan: So, of the latter. So basically, we take three or four pictures of the same patch of sky, And Then we stack them up. We align all of the stars and galaxies, uh, things that are going to be in the persistent across the images, and then things like the cosmic rays, which are just, you know, stray particles that got in, will only appear in One of those exposures, and so we say, Okay, this thing only appears in one exposure, it's a cosmic ray, we'll just, you know, draw a little sketch around those pixels, and then we won't have the data from this exposure, but we can fill it in with what's in the other two or three.

0:47:15 Thomas: Hmm. Oh, OK.

0:47:46 Thomas: Interesting.

0:47:46 Bryan: Yeah. That's the first stage, is all of the cleanup. Then the next stage is the object detection, so this is a very common problem. We've got standardized tools for this. And this brings me to another case of astrophysicists coming up with really bad acronyms for naming things. So I'm going to ask you to guess this one, see what you come up with, and if it's as bad or worse than what we have to work with. So the tool is source extractor. So given those two words, what would you name that tool?

0:47:47 Thomas: OK, cool. OK. Hmm. Just from two words. Boy, how bad can an acronym get from two words? Source X. I mean, I would have just said like, oh, Source X or something. S Source. I don't know. Two words.

0:48:23 Bryan: Yeah.

0:48:32 Bryan: Source X. Okay.

0:48:33 Thomas: There's just be S E. E. Wow.

0:48:34 Bryan: Yep. Either of those would be perfectly fine. So what ended up happening is that we took the S from source and the extractor from extractor, And We get sex tractor. Yeah.

0:48:45 Thomas: Oh, come on. Oh you folks. That's

0:48:49 Bryan: Yeah.

0:48:49 Thomas: Stupid.

0:48:51 Bryan: Yeah, it's one of those things that you can tell astrophysics was a bit of a boys club and that you know, no one checked this and said, okay, you've got to rename this, but this Tool Has

0:48:57 Thomas: Yeah. That, that feels intentional. That's not a, Oh, Oh whoops. I didn't know. Come on. This is, that's nonsense. All

0:49:02 Bryan: Yeah. No. Well, It's definitely intentional in that when you're, you, this program, the command for it is just sex. Okay, you, you definitely did This on Purpose. Purpose.

0:49:16 Thomas: Right. Grow up astrophysics. Come one.

0:49:19 Bryan: Yeah, yeah, so we've got this tool and we're just going to call it S-Extractor because It's It's Like the planet.

0:49:23 Thomas: Yeah, we're not playing by those rules.

0:49:25 Bryan: Um, That we Know it was meant to be pronounced a different way, but we're calling it Uranis to avoid the dumb jokes, and maybe someday we'll grow up and name it something sane like your rectum, But So we've got this tool called S-Extractor, which just automatically looks over the images and looks for clumps of pixels that show a distinct shape and identifies them as objects. We match up to other observations.

0:49:26 Thomas: Yeah Yeah. Yeah. Yeah. The Futurama joke.

0:49:56 Bryan: Euclid isn't the only thing looking at the sky. There are tons of other surveys doing it. So we try to pull in other surveys looking at the same spots and match up, see if we can say, oh, this object was also seen by this other survey. So we can get in some extra data there. And then we try to compile a catalog on each object we identify to get a good Idea of if it's a star or a galaxy. And if it seemed by something else, do we have a good idea of how far away it is? Then once we have all of that, it comes into our group, which is the shear analysis group. So we've got this object, and to give you an idea of the scale, it's going to be a few pixels in size. So this could be anywhere from three to ten pixels at the extreme small limit.

And so you've got a blob of three to 10 pixels, and you've gotta say without bias what shape it is. You can think about it, okay, we've got three pixels, it's a triangle, but yeah, have to come up with some algorithm that will as unbiasedly as possible tell you what the shape is. And this is accounting for the facts of one, the pixelization, and two, the PSF, and three, any noise that comes in, since we're looking at the extreme edges of detection.

So we've got a task at hand, and there's a lot of scientific work trying to figure out what's the best way to measure the shape of something this noisy,

0:50:19 Thomas: Hmm.

0:51:29 Thomas: Wait, why is it only three pixels? Or are you just saying that's all

0:51:32 Bryan: We want to go as deep as possible and get data on as many things as possible. So, yeah.

0:51:32 Thomas: You have? Oh, right. So inevitably there's going to be something that's like the smallest amount of thing you can even look at reasonably. Okay. Gotcha.

0:51:41 Bryan: Yeah. And I think three pixels is pretty much the limit. The limit. We're not going to... one pixel is just noise, two is chance, three is okay, there's a good chance there's an object there. But practically speaking, we don't get too much reliable data from the three pixels, but around 10 pixels, yeah, we have to try to measure those.

0:51:45 Thomas: Jeez. Yeah, wow.

0:52:03 Thomas: Jeez. That's like, that's, that makes N64 graphics look really, yeah, really. You just got this this little tiny blur and you got to be like, all right, what's there? That's

0:52:07 Bryan: Yeah... this is Worse than Pong, really.

0:52:14 Thomas: Wow. Yeah. Yeah. That, that sounds really difficult.

0:52:15 Bryan: Yeah. Ways we try to do it, there's one way is just mathematical, like we want to calculate the shape. So we take, okay, what's our shape calculation formula, how do we define how you calculate shape, okay, throw that formula at those pixels, see what we get out, and then do some corrections for the effect of the known PSF.

That's one way to do it, which is very fast to do, and And it has worked reasonably well in the past, but as we get into smaller and smaller and more and more difficult images, and need to get the bias lower and lower, it turns out that it's hard to get rid of all of the bias in this, just because we're working with things that are so damn small. All. So we've come up with other ways to look at it.

So we use is called a forward modeling approach. And it basically says, okay, we imagine that there's some galaxy out there that has some shape, and then we run it through all of our knowledge of the telescope, like what's the PSF, the uh, pixelization, with how large it is, and how's that galaxy going to look? And then we we compare the two, and obviously our first guess is going to be way off.

But by comparing and testing a few, we can get an idea of in which direction it's way off, and then we shift off in the direction that will get it closer to this. So essentially we just keep simulating many galaxies over time, getting closer to one that looks like what we simulated, until we finally have one that said, that, okay, this theoretical galaxy with the PSF we know, with the pixelization, could potentially generate an image like this. So we're just going to take the shape of that theoretical galaxy and record it as our measured shape. Do you follow along with what I'm saying there?

0:54:10 Thomas: So I mean, it does sound like a bit of magic, you know, it's like taking really a tiny amount of data, but by dialing it in and I imagine it's also because like there's only so much it could be. I mean is that part of it? Like you have a lot of constraints in terms of like what you're likely looking at. So that helps just statistically to try to refine your guesses as to what it is, so

0:54:22 Bryan: Yeah.

0:54:31 Bryan: Yeah, exactly. So most distant galaxies are going to be either what we call an elliptical or a disk. And so an elliptical just looks like some circular blur stretched out in some random direction. And a disc, you imagine a plate floating up there and it's tilted at some angle relative to you, maybe with a bulge in the center. And so you

0:54:54 Thomas: You gotta sex track the bulg, um So But like that's cool, what I'm always curious, I think I kind of asked you this last time maybe, but what would happen if, if you were indeed truly looking at something kind of unprecedented, how you were to guess, like how, how missable do you think that is? Like something truly crazy, would, would.

0:54:56 Bryan: Yeah, there you go.

0:55:12 Bryan: Happens all the time. Time. The ellipticals and just, yeah, well, we actually just have a name for it. We call them irregular galaxies.

0:55:15 Thomas: Yeah. Like a triangle galaxy. Uh -huh.

0:55:23 Bryan: And these happen all the time because galaxies every once in a while run into each other. Now, the stars don't actually hit each other, we talked about that in the previous, but the gravity affects them and it'll distort them and you'll get this tangled mess of a couple of galaxies. Like you can imagine that what we think is one object is actually two objects, one right in front of the other or they each other and shred up one, and so we get what we call these irregular galaxies. And for we're trying to model it, in these cases we don't include irregular galaxies, as the number of possibilities is basically infinite for what irregular galaxies can

0:55:28 Thomas: Yep. Yeah. We talked about that. Yep.

0:55:43 Thomas: Mm -hmm. Yeah. Right.

0:56:02 Bryan: Look at. So some are going to be missed, so what we try to do is we do do everything possible to fit it with our known simple models, which do cover the vast majority of galaxies. And then if in the end, the resulting image isn't close enough to the actual image that we see, we flag it as could not get a good fit for this image, don't include it in the final data set.

0:56:11 Thomas: Mm -Hmm.

0:56:28 Thomas: Oh, okay. Interesting. Interesting. So it does that mean that when it comes to irregular galaxies, stuff that's smushed or weird or whatever, if they're far enough away we're probably just not gonna confidently say what it is? Do they need to be closer and more pixels to work with to be able to be more confident?

0:56:45 Bryan: It depends on what the purpose is. So for what we're working on here is trying to do gravitational lensing for which we want to know the shape of it. And in particular, we want to have reasonable confidence that it was one of these spiral or elliptical galaxies. It lensed a bit. It's close enough that we can make it out, often it isn't suitable for lensing at all, as there's too much data in it, and we'll get too much information on the internal structure, and all the tools we throw at will actually detect it as multiple objects. So like a spiral arm might be detected as its own object, the bulge as its only object, and the other arm as its object.

And so it would be picked up in our machinery as three different objects. And so a part of what we do is actually we mask out anything that's too close to us as being not useful, as remember the point of lensing is to figure out what's happening in the space between us and the source galaxy, because light is bent, and anything close enough that we can clearly make out what it is is also going to be close enough. There's not going to be much of interest between us in it, so we don't really care too much what its shape is, as what's between us and it is basically nothing of interest. Yeah.

0:57:05 Thomas: Right. Right.

0:58:06 Thomas: Just feels like so many variables, you know, it could be irregular, could be lens, could be whatever. That's uh, sounds like a difficult job is there a lot of disagreement, like, or revision or whatever, in terms of those things, I

0:58:08 Bryan: It is Yeah. Oh yeah, it does. Now, Oh yeah, all the time. Like, oh, what actually did is we had multiple um competing methods for trying to calculate the shapes of galaxies. Uh, we started out within Euclid, we had four different methods that we were working on. The one I was in charge of, at least in part, was I took the old method and just retrofitted it to run on Euclide. And this basically just uses a baseline to show that the new ones were better.

So I just took the old one and showed, okay, here are the crappy results we get with the old ones. Now look at the the new fancy results we get with the new methods, and we had three that we were working on, um, eventually decided on the forward modeling approach I described to you, uh, go forward with for our final, uh, measurements. But yeah, those methods aren't the only ones out there.

There are plenty of people outside of Euclid that have their own preferred methods with benefits and drawbacks, and so there is disagreement about how to best get these shapes. Yeah, so yeah, much information here that

0:59:17 Bryan: There's no firm place where the science actually ends. There's only a place where the conversation ends. So for today, the conversation ends here.

0:59:17 Thomas: Yeah, sure. Yeah, well, once you remind us, your overall, you know, you were trying to teach us, and kind of where we are, maybe what, what more there is to talk about.

0:59:35 Bryan: Overall lesson for this is first just to give you an idea at a high scale of how much work goes into every aspect of this, that, you every know, little bit of a telescope like this has to be carefully considered, considered. Um, then we have to have careful knowledge about all of it to have as perfect an understanding as possible about everything that happens to the light once it hits the telescope and everything that happened to the telescope over the course of the mission, so that we can account for all of these effects, and then this all comes together into um pretty boring black and and white images. Yeah.

1:00:16 Thomas: In The end this kind of sucks, not gonna lie. No, but it's such good data.

1:00:19 Bryan: No. No, It's very good data, it just doesn't look like much. Uh, eye, that's not the purpose of Euclid, it isn't trying uh to get super deep, it's trying to go for wide high quality data. And then what's really going to be interesting is once we get this data from a third of the sky, in total, um, it together, run the calculations on it, to figure out, going back to the previous episode, we want to figure out how lumpy the universe is, at what point in time it got that lumpy, and then we can

1:00:28 Thomas: Yeah.

1:00:52 Thomas: When somebody left it out, you know, didn't put it back in the fridge. Exactly.

1:00:53 Bryan: Yeah. Oh, Yeah, all these stray universes left around, nobody's drinking them up, and hopefully we can get an idea from how lumply it is, how likely we are to be drunk up by some interdimensional god.

1:01:00 Thomas: To get food Poisoning. Yeah. Yeah, as you said, the never ends. And it is the resolution at which you could explain this stuff rivals the resolution of the Euclid Space Telescope. So we'll call it there for now. I'm sure that won't be the last we hear of the Euclid Space Telescope. And you said this is set to go up pretty soon.

1:01:06 Bryan: Yeah, yeah. Yeah. There we go, uh, in just a few months now. It's going, going to be going up in this summer on SpaceX. So cross your fingers, Elon Musk doesn't cause that company to fail.

1:01:31 Thomas: Wow. SpaceX. Oh, don't make me root for anything. Elon Musk is involved with, come on. Oh well. He's, he's gotta be pretty hands -off, right? I mean, I feel like there's probably like really smart people who work there who were like, I hate this fucking guy, you know, and they're, but like, you Know, and coincidentally seems to be the most, the most amazing of the companies.

1:01:43 Bryan: Yeah, I know. It wasn't my choice. Yeah. Yeah he's gotta be. SpaceX seems to be the one of his companies that he's the most hands -off with. So it should be all right. Yeah

1:02:01 Thomas: It seems like to me. Um, really really cool, though. Is it the kind of thing I know that, you know, with James Webb, again, it might be the difference in the political situation or whatever. I felt like there were a lot of false launches. There's a lot. Also, weather, I think. Yeah, that was.

1:02:02 Bryan: Yeah you can never Account for weather. There's always a chance the final launch will be delayed because it's too windy that day, so yeah, there's no guarantee there, it will go up on any specific, uh, until

1:02:27 Thomas: You know much about like the launch windows on that? Like, let's say, oh, that day it's windy. Does that mean I will go tomorrow or does that mean now got to wait till next year?

1:02:35 Bryan: No, it's not necessarily tomorrow because you've got to account for Uh, the motion of the earth and the moon and that, well, you just send it up one hour later, the earth has already rotated a good chunk of the way, and so you'll be sending it off in a completely different direction.

1:02:38 Thomas: Right. Yeah. That's Yeah. Right. Right. That's exactly what I'm asking, so like, do you have any sense of like, when's the next, what, you know, our windows once every what, month, year, week?

1:02:57 Bryan: It's... they're usually... well, it depends on what you're doing, but it's usually weeks apart, so it probably wouldn't be that much of a delay if there is a weather delay.

1:02:58 Thomas: Hmm. Well, We'll keep keep an eye on that. All right, Brian, uh, as usual, thanks so much for coming on and explaining some more science to us and I would like to talk more about models of the universe and whether they're infinite and all that stuff, but I don't know if that's like your specialty or anything, but that's the stuff that I just find so interesting. So maybe a later date.

1:03:13 Bryan: Yes, thanks for having me on.

1:03:24 Bryan: We can definitely find time to talk about that, yes.

1:03:26 Thomas: Excellent. All right. Well, until next time.

1:03:29 Bryan: Right, thank you.