Dr. Michael Blanton on Open Data, Galaxy Surveys, and the Future of Astronomy

Intro

0:00Hello internet, this is your captain speaking, Lester Nare. We have begun preparations for our return to regularly scheduled programming with our resident PhD, Krishna Choudhary, expected back on the pod later this June after some well-deserved family leave. But today, we have our second interview in our ongoing collaboration series with Carnegie Observatories. In this interview, Krishna sat down with Michael Blanton, the new director of the Carnegie Observatories for a conversation about open data, the Sloan Digital Sky Survey, the future of astronomical data science, and the

0:41mystery of why the universe's most massive galaxies stop forming stars. And one quick announcement before we jump in. For those of you who are here local to Los Angeles, Krishna will be giving a talk at Exploring Physics at UCLA, hosted by UCLA's Physics Outreach Organization, Continuum, on Saturday, June 6th at the Fowler Museum. His talk will run from 9:30 to 10:30, and we will include a registration link in the description. So, if you're in the neighborhood, come stop by. I'll be in the audience. It's sure to be a great talk as we warm back up to our regular

1:23episodes. But, as always, we are going to learn about the science from the ground up today because this is From First Principles.

Michael Blanton joins the show

1:33>> [music]

1:47>> I'm joined today by Michael Blanton, the new director of the observatories here at Carnegie Science. Michael, thank you so much for joining us. >> Thanks for having me. >> Um you know, I wanted to start by taking a trip down memory lane because both you, me, and Lester were um part of the institution at Princeton University. You got your PhD there in Peyton Hall. Um I was in the neighboring Jadwin Hall, and I was always very jealous of all the astronomers for having such a beautiful building, whereas we got, you know, whatever monstrosity um Jadwin Hall was.

2:28So, one question from one Princeton alum to another Princeton alum, always um how often do you go back for reunions? >> I never go back. >> You never go back. >> a I was a PhD student there, and and so we existed in a slightly different culture than the undergraduates. >> Yeah. >> Uh you know, eating clubs for us. >> No, yeah, but Peyton Hall was pretty close to all the eating clubs, so that was I I I always You know. >> walk by them on my way to Hoagie Haven, but >> Oh, boy. Um I'm sure you didn't think very highly of the undergrads there, but we probably uh >> Actually, they were great. I um I TA'd while I was there, and and they were amazing undergraduates, so.

3:10>> Well, just as a pitch, you know, um there are graduate student alumni uh reunions here at the same time, so you should you should you should uh >> Yeah, they they they I did go to the reunions when I was a PhD student there. They were really great parties, but I haven't been back for any since then. >> Yeah, of course. Um well, so you got

Sloan Digital Sky Survey and open data

3:29your PhD there at in astrophysics. Um one of your advisors was Jeremiah Ostriker. >> Yeah. >> And he was seminal in starting, or at least initiating, the concept of the Sloan Digital Sky Survey, which was this monumental effort to map out a large swath of the universe. And um one of the things that I've read about the SDSS is that it completely sort of changed the culture of astronomy from, you know, now you guys were just putting out the data for free on this newly formed internet. Could you talk a little bit about what astronomy was like back then and how it's changed today?

4:11>> Yeah, so so Jerry was one of my two advisers. Um, and indeed at that at that time he was putting together the the Sloan Digital Sky Survey. Um, you know, at at you know, if you look back at astronomy in the, you know, before around that time, it was the mid-90s, you know, a lot of the data was accumulated in non-digital forms. So, there were still people collecting data on photographic plates. >> Wow. >> Like glass plates that had emotion on them. They had uh uh and digital detectors had been

4:51introduced. Um, and uh other electronic detectors uh before, you know, uh before that. Um, but they they really hadn't been as large-format. You couldn't do like a large like if you were imaging the sky, you couldn't get like a really big CCD to actually do that. Um, and and so for that reason a lot of things were still photographic. Um, you know, the uh the you know, starting around 1990, it became possible to conceive of things which were bigger and digital. And once

5:32you are doing something digitally, you are then able to share it in in a much, you know, much you know, much more broadly, much more easily. >> Right. >> Um, and so a few people recognized that that was going to be a possibility and for example the people at you know the Hubble Space Telescope which obviously was sending data down digitally. >> Of course. >> Uh they they had um they had a lot of uh you know data distribution uh to the community um and SDSS was this other big digital thing that was happening and partly

6:13in order to help fund the survey and motivate the National Science Foundation and the Sloan Foundation to fund it they promised somewhat uh uh somewhat rationally maybe to make it all public and freely available. >> Yeah. >> Um and and that was very new just because both because of the technology but also just because of the culture there and I there was a genuine concern among people that like oh you're really giving away the store here you'll get scooped. >> Yeah, yeah. >> Uh which is >> I remember that. >> No we which is a term I normally associate with you know you know

6:53you know characters in the front page or something or rather than scientists but but but some scientists are afraid of being scooped. Okay. Um and but SDSS really from the beginning had a very different philosophy of of eventually making the data fully public. This I was a graduate student at the time I was doing theory. >> Right. >> So I wasn't I wasn't part of conceiving of it that way but when I heard about it it seemed like so intuitive like that's the right thing to do you you know you make the data available you make it uh uh so any astronomer can get to it and you're going to get a lot more out of the data. They'll people will be able to collaborate freely they will do things

7:36you know you hadn't thought to do um and there's just too much science with a data set like that for everybody to do. Um I think what's been surprising to me over the years is that so to do that right, you have to really spend enough money to make it good. >> Mhm. >> I think the genius there was Alex Szalay at Johns Hopkins who actually, you know, made the the SkyServer and later SciServer which which really made things accessible uh to everybody to like middle school students and >> Right. >> as well as astronomers. Um and you you have to spend the time and

8:17effort to do that. And I think what's been surprising to me over the years is you have to keep telling people that lesson. People still have an instinct to keep their data close and to not spend the effort to make it freely available. Um and uh and that

Rubin Observatory and public alerts

8:35sort of argument I thought that that argument was won once and for all. >> Right. >> Uh but it was not. It has It has to be re-argued, you know, almost every project. >> Yeah. >> So that was surprising to me cuz it seems so natural. I don't know if that answered your question. >> No, totally, totally. And I think you know, modern astronomy has definitely changed a lot, right? Because I mean, with the Vera Rubin that's coming online, that's very similar in that sense. The the data's going to get fed to whatever server very, very quickly as soon as the telescope takes that photo, right? >> I uh yes. So the So for Rubin, you know, as soon as they you know, a-

9:18you know, as soon as they get the data, they will be sending out alerts publicly. Right. Sometime this year they'll probably start. Uh and those will be fully public. And then the data itself eventually becomes uh uh public and and will be distributed to the the partners. >> Yeah, yeah. I think that that's a really cool thing about um astronomy that I think other sciences have not totally gotten to where astronomy has, you know, in terms of just public distribution of data. I think that's that that's something that other sciences can probably learn from. >> I I I think so. It's helpful, and this is one of the reasons I think that people who work in data science and

10:00machine learning and what have you, I think really like working in astronomy as >> sort of a sandbox. >> Yeah, I mean, it's a sandbox data, I mean, it's like our livelihood. I I But um I think I think the reason they they like that is actually there is a lot of public data, and there's no privacy issues. >> Right. >> So, it's it's sort of uh There are many, you know, it's like there are many sciences where it's a legitimate problem if the data becomes available. Like in the social sciences, it's not so easy. >> Totally, yeah. Um in the neurosciences, too, if you're dealing with human data, but you know, there's a place called Allen Brain Institute in Washington. They're publishing a lot of, you know,

10:41mouse and um rodent data freely available, and that's sort of changed um >> The mice are unlikely to sue. >> Yes, yes, very much so. Um you know,

NYU Value-Added Galaxy Catalog

10:52keeping on that trajectory, your career has also been one of making large data sets available to people, and have those data sets be such that it's easier for astronomers to analyze. Right? And one of the things that I'm referring to here is the NYU value-added galaxy catalog. >> Yeah. >> Um it sounds like a venture capital thing, you know, what what is the value you're adding to my portfolio? But um if I were to ask you, >> Yeah, yeah. >> what is the value that you added to the SDSS with >> Yeah, in retrospect, we would have handled that name differently, But,

11:32>> [laughter] >> Well, you're coming to California, so >> Yeah. Yeah. >> It's the right wording here. >> I mean, really in that case, um one of one of my personal obsessions about doing these large uh uh you know, analyzing the and taking these large data sets is that for a lot of things you want to do, you don't only want to study you don't only only want to say, "Oh, I've detected these galaxies. They have these red shifts, whatever." You also need to know exactly what you did, like where you could have seen a galaxy. If there was a galaxy there, would you have seen it? Right? Because you can only see the

12:14galaxies down to a certain brightness. >> Right. >> And that means for a galaxy that is of a certain luminosity, like it's intrinsic brightness, that means there's a distance out to which you can see it. And so so when you observe the sky, you're mostly seeing nearby things and fewer faraway things. >> Yeah. >> And you know, the things we were trying to do uh with SDSS at that time was map the universe on the largest scales. And if you just took that map, you would be like, "Wow, everything's near us." >> Right. >> Uh and you have to correct for that and correct for the fact that we looked in this part of the sky and not that part of the sky. So so most of the work of

12:55that catalog was very carefully, you know determining what parts of the s- universe we we we uh had actually been able to see things so that we could then do things like measure the statistical correlations between galaxies over very large distances. So that was that was sort of the main reason for that catalog. It was sort of uh so that statistical analyses, you know, had a had a firm basis. >> Yeah, that that totally makes sense. And I might be going off on a limb here, but you know, one of the things that I hear in cosmology all the time, um at

13:35Princeton I actually took an undergrad course with Neta Bahcall >> Oh, yeah. >> on cosmology. It was great. She's She's an amazing professor and I learned a lot. Um and you know, one of the things that we learned in that course that was kind of a central tenant was that the universe is isotropic. You know, whichever way you look, it's about the same. Was your Was your, you know, value-added galaxy catalog part of that conclusion? >> So So, yes. Uh

How uniform is the universe?

14:06I think that the strongest evidence for that comes from the cosmic microwave background. >> Oh, okay, fair. >> Where it's where it's just very clear that like that part of the sky and that part of the sky, which have never talked to each other, are >> the same temperature. >> Yeah. >> Um So So, I think that that is you know, is the strongest evidence, but it is true that um with with the largest volume sample of SDSS, we were able to do that for the galaxies as well. We could answer the question like how homogeneous is the universe. Um and uh and and we

14:47did we did have a a quantitative paper, you know, based on the volume we surveyed, which which showed that it's very very uniform. >> Yeah, it is very very uniform. >> on those large scales, right? On the small scales, everything's clumped, but once you're at hundreds of millions of light-years or a billion light-year scales, then everything's [clears throat] pretty much the same anywhere. >> it's like soup. >> Yeah. >> Yeah, very cool. Um So, now I want to

Why Carnegie Observatories

15:15talk about your transition to Carnegie. You've got a new job now. Congratulations, by the way. >> Thanks. >> Um you left New York for the best city on the planet, Los Angeles. >> Okay. >> I mean, you left your native >> out, but you know, not to us. >> So, I grew up here and uh you know, I've always loved it. So, welcome to Southern California. This is an institution. Um Hubble walked these halls, you know. Why why make the move? What about Carnegie brought you here? >> The astronomy. >> Yeah. >> Um so, it's in a department a department of physics at NYU. Um and one of the things I've I've

15:56even in this first month is just really clear is is just uh just every morning there, you know, uh there is such a range of experts here talking about every single aspect of a of a astronomy. You know, yesterday someone's talking about very large-scale structure. People are talking about exoplanets. It's really like that. So, one of the main draws for me was just that environment where where uh where I'm learning something new every day from the people around here. Uh and

16:37and so so, that's actually the fundamental thing. Um another thing is just simply the resources in terms of the engineering resources, Las Campanas Observatory is an amazing facility. Um and uh and and SDSS actually partnered with Las Campanas. We actually um >> Oh, yeah. >> Now, uh you know, uh in SDSS 5, they're completely using the 100-in telescope. Um and that's that's a uh you know, uh so, I've had a you know, a long you know, a decade or so of experience

17:19working with Carnegie already through that. >> Right. >> Um, and then in terms of the job itself, like a director job, I was I you know, I started in SDSS as like, you know, writing software, which actually kind of still do. >> Right. >> But in in a previous phase of SDSS, I was the director. Uh, and I sort of learned at that time that I I kind of liked enabling science. I got a satisfaction from making it possible for people, as you said, get at their data, but also get the science they wanted done out of the project.

18:00And so it's an admit like that was an administrative and leadership role, uh, but very close to the science. So like not like no offense to any deans, but not like being a dean, like like really like very close to the scientific decisions. >> Mhm. >> And and I think and this job being director here, I I think has that same flavor. You're very close to what people are doing and the you know, the sorts of decisions they have to make that that require scientific judgment. Um,

Why astronomy is all connected

18:35and and so I like that place where it's a leadership role and you're really it's enabling science, but you're you're sort of close to you know, what's actually happening. >> Right, yeah. I mean, just before we were filming in this room, there was I think a journal club meeting something. >> team. >> Yeah, they were discussing the Bullet Cluster data from the James Webb studio. And I was just eavesdropping and it was such a cool conversation that you know, everybody was just gathered around talking about one paper. So it is an incredible institution. >> Well, the this is amazing thing about astronomy, which which is different than other fields is that well, I don't know how different it is from other fields, but I think it's different is that you know, when you're an astronomer you kind of have to know

19:17all the bits of astronomy. Like they actually all end up being relevant in some way. Uh, so it is relevant to the people who study stars what's happening in the Bullet Cluster. Like that turns out there are enough connections to like how galaxies form and then how stars form and the elements that are in stars that that those things are connected enough that you really have to you know, have some knowledge of what's happening across the field. >> Yeah, so it's not just like dark matter guys or girls looking at the Bullet Cluster. It's like everyone. >> Yeah, well, no, no, I think it's certainly very relevant to to the people

19:58who study uh, galaxies. But then that is connected to the people who study stars, which is connected to the people who study planets and and and it's very hard to you know, you're kind of expected as an astronomer to have some handle on all of that. Uh, and and that I think is and partly it's because it's you know, there's only whatever tens of thousands of of professional astronomers in the world. So it just is a small field. But um, but that I really like about it that you're you you sort of have a connection to almost every astronomer that you meet. >> Right. Yeah, that makes sense. Um, so you've been

Blanton’s priorities as Director

20:36you're you're very new director here. >> Very new. >> What are you looking forward to in your tenure here? >> So there's um, you know, the the things that are going on here uh, the really the main observational event right now is is Magellan, two six and a half meter telescopes. Um, and uh, and the SDSS program at the DuPont. Uh, and what I think, you know, our challenge, one of our challenges, is, you know, with any, you know, any large telescope like this, really

21:18any telescope, you want to make sure that the things that are happening at it are continue to be cutting edge. So, I think one of the things we really have to to figure out is the strategy for for for keeping the telescope functioning. It was built in the '90s. It's got all of the, you know, it's got some '90s technology still in there. There are things like that which are which which really have to be done. They're sort of more on the technical, you know, uh, infrastructure side. Right. Uh, but then there's also there are new opportunities for instrumentation now. There is there's instrumentation that that you could build better now and be able to look at

22:01fainter things, look at things over, you know, wider swaths of sky. And I think that figu- you know, in the in the in that those sorts of time scales, I think we have to figure out how to build those instruments. And and one of the things about that is that uh, and I think this is borne out by some of the successful efforts, recent efforts, is that you really have to do this collaboratively. >> Right. >> Like, you know, Carnegie has to be working with, you know, and this this this is part of like the way Carnegie has worked with SDSS, which has brought a lot of new instrumentation and energy to certain parts of of

22:43Carnegie. Um, and I think this is true

Astronomical data science and AI

22:46at Magellan as well. The 6 and 1/2 meters is, I think you have to be working with other other parts of the community to build the things that are cutting edge. You don't do it alone anymore. So, that that's one aspect. I think there's another aspect which is that I really think that you know we are sitting on a pile of data. It's a huge pile of data and and facilities that you know that can create them and I think that there there's a lot of work in uh you know I sort of think of it as astronomical data science but that would

23:27include everything from from soft you know telescope software to to you know data reduction software to machine learning to you know AI applied to to to the data like I think that that is something where I think Carnegie should have like a really strong position in and so that that's something I would really like to to see happen. um >> Yeah, do you have a do you have any

Photographic plate archives

24:01plans with all the photographic plates in your basement? >> I think no the answer to that is no. >> Yeah, I mean it's always a hassle. >> So so there there there are program like Harvard has a program for their library where they they're digitizing all of these things. I I think that is in principle interesting. I think you have to I think you have to do it in a science driven way. I think because it's such a big problem. I mean you've seen >> Yeah. >> Yeah, you know there was all sorts of things all sorts of different data all sizes and shapes. Um and so you really would you want to approach that with I really want to

24:41study X and I'm going to do something I think that's the way you approach that. >> Yes and and for what it's worth I mean any astronomer can request Carnegie for like a specific plate and things like that right? You have some way of doing it? >> I don't know the answer to that question. That's one of the things I >> Yeah. >> I should probably know. >> Okay, well, I might be misremembering, but last time when we got a when we got a tour, they were like, you know, people come in and they say, "Oh, I want to look at this part of the sky." You know, and there was some somebody who could >> we say yes to that. That would be great. >> I think I think you do, but maybe I don't know. Maybe >> Yeah. Yeah, I know. Absolutely, we should. Um

25:22and there, you know, there are definitely things in astronomy where people have studied this star for >> Right. >> hundreds of years and and and this is a hundred year period where uh especially if it's something significant enough that people are doing that. Yeah, there probably is informa- information in our archive about it. >> Yeah. >> Um >> Definitely. I mean, your archives here are just spectacular. So, um one of the things that you talked about was that, you know, as director, you're not like you're not someone like just pure administration. You're making scientific decisions. Um you were the director of SDSS-IV.

Scientific decision-making at SDSS-IV

26:03>> Yeah. >> Right? So, could you give an example of like a scientific decision that that you, you know, it was the buck stops with you and you had to make that decision? And what what what were the things that you weighed when making that decision? >> So, um I think there >> And it doesn't have to be from SDSS. If anyone here >> Yeah, I I think SDSS probably has has the examples I could most easily talk about. You know, I I I think there is I Just tell me if this is too far in the weeds, but you

26:44>> We love the weeds on this podcast. >> the weeds. Good. So, in SDSS-IV, we had two programs that were using what we call the dark time, which is when the the moon is is is new. >> Oh, okay. So, on the other side of the earth. >> Well, it's on Yeah, it's on uh it's it's on the >> the day side >> the sun, right? So, it's it's when the sky is dark. Right? So, so, one one was called Manga, which is another naming uh Yeah, it's spelled anime. Yeah, spelled the same way. Makes it hard to >> What was the What was the full form? >> Uh the mapping nearby galaxies at Apache Point

27:25Observatory. >> Okay. >> Yeah. >> That's Yeah, that's okay.

MaNGA vs. eBOSS

27:30>> But, it was observing nearby galaxies. >> Great. >> Um and then there was eBOSS, which was like the last red shift survey we did. >> Another great name. >> Yeah. Yeah, eBOSS Yeah, that was the extended BOSS survey. Anyway, so anyway, retrospectively. But, um but what was happen What we had to make a decision between this, you know, how much time we were spending on each. >> Right. >> And you know, there was a question of uh the competition for each of these two. And basically, Manga actually, for a data set like that, the way the the amount of information we're getting per galaxy uh the number of galaxies, it actually this program ended in 2021,

28:12but there's still nothing that's comparable to it even now. >> Right. >> Whereas with eBOSS, it was our last red shift survey, and another red shift survey called DESI was starting up near the end of when we were uh ending, but not but you know, the it was going to be a small overlap, and our eBOSS team was like you know what? Like, we all want to work on this other project. >> Yeah. >> You know, and we we were really want to close down eBOSS sooner, but we want to reach our goals. >> Yeah. >> And so, the decision there was sort of let's bring forward eBOSS and just spend, you know, I forget exactly, but spend like a whole year

28:54just doing that. Don't do any Manga. And then wait for Manga to do in the last year of the survey. I'm getting the exact timing wrong, but I know so what back. There's a whole report to the Sloan Foundation about it. Um but uh but but that was a decision where it's like we're going to put your data taking on hold so we can these >> people done. >> And then we'll go back to you, which is quite risky because at that time we didn't have enough money to finish the project. >> Oh, wow. >> Yeah, yeah, yeah. So >> Okay. >> Yeah, yeah. So it was um we didn't have enough money to finish the project until maybe 18 months before we finished. >> Wow. >> But part of that So that's a scientific decision. It's like do you

29:36prioritize this thing and put the other thing at risk? And then you you sort of say, "Well, we'll give you a little more time if you'll take it later." >> Yeah. >> And and so you work it out that way. And it also became uh a a selling point to the DOE to give us a little more money, the Department of Energy, to give us a little more money to finish eBOSS. >> Right. >> Uh uh and so so that's an example where it's uh a scientific decision because of the timing of these projects. >> Yeah. >> And but it's somehow intimately tied to the funding decision. Like that only works if you can actually get a little money that helps you close your budget.

30:17And so that's an example where where it's like it's not just an administrative decision or a scientific decision. They're they're intimately connected. >> Right. And just to be clear, these are two programs that are apps king different questions scientifically. >> Totally different. They're trying to use the same telescope. >> They're using exactly the same telescope. >> Right. And and eBOSS is your the red shift? >> Yeah, it's >> So that's the one that's just like charting galaxies far away. >> mapping the universe on the largest scales >> Right. At the time. Yeah, at the time. But I guess DESI was coming on and I think DESI's at in Tucson, right? In Kitt Peak? >> Yeah, it's at Kitt Peak. >> So it's like nearby. >> It's nearby and it was the same same of

30:57people. Like the team of people boss eBOSS were were the people a key part of the team that had proposed DESI. So >> And they'd already gotten time so they're like let's finish up. >> Yeah, they were they they were sort of you know, we're going to we really want to move on to this other thing and and and you know, because for good reasons. And so so that was their motivation. Okay.

MaNGA and galaxy spectroscopy

31:24>> Very cool. And then I guess MaNGA was nearby galaxies. >> Yeah. >> What was the what was the question cuz I I I imagine with red shift you're trying to answer in a Hubble's constant expansion of the universe. >> Yep. >> Cosmology at the biggest scales kind of thing. >> Pretty much. >> What was MaNGA's prerogative? >> MaNGA was and MaNGA's tagline was the galaxy survey for people who love galaxies. >> Okay. >> So so it was getting so it was this much smaller number of galaxies. I think by the time we finished the SDSS red shift surveying at the end of eBOSS we had maybe somewhere between three and four million galaxy red shifts. MaNGA observed 10,000 galaxies. It was much smaller number.

32:05But for each galaxy you are observing not just you know, the the the red shift survey just takes the spectrum of the center of the galaxy. >> Right, it's like the mean. >> It's just get one red shift and get and you get more information. You get something about the overall age of the stars, the metallicity of the stars, something about the gas in the galaxy. But MaNGA you had these fiber bundles where you put the whole fiber bundle they're not this big though, like that big. >> Yeah. >> But um >> Optic These are optic fibers. >> Optical fibers. >> Not the string theory fiber bundles, just >> That is, yeah, there's no Lee algebra here in the creation of these fiber bundles.

32:46Um the uh, the these fiber bundles are pieces of glass, um, and and and they're put on the galaxy in the focal plane of the telescope, and we would do 17 of them at a time, uh, and get information across the face of the galaxy. >> Oh, very cool. I >> So, you could get a rotation curve, you could get, you know, the gradients of the gas distribution and the metallicity distribution. >> Right. >> Uh, much more information about, uh, you like if there was an active galactic nucleus, which is like a supermassive black hole at the center of the galaxy, you could get that separately from the rest of the galaxy.

33:27>> Yeah. So, before there was just one spectrum, and the entire galaxy, all of its spectrum was coming through, and you're getting this thing. >> Yeah. >> Now, you're sort of like pixelating a galaxy and tiling it. >> Yeah. >> Because the spectrum over here, because it's moving, you know, the it maybe it's rotating this way, is different from the spectrum over here, is different from the center if there's some black hole that's acting up. >> Yeah. Very cool. >> I see. Yeah. That's that's that's super fascinating. Um, I wanted to wrap up with two questions. >> Okay. >> One was, you know, with your with your tenure coming up in Carnegie, what is like one mystery of the universe that you personally are super, you know, curious about, and you

34:08hope that your efforts can can lead to an answer? >> I mean, uh, there are many. >> Yeah. If you were to pick, you know, one that >> If I'm going to pick one, personally, >> Personally. >> Yeah. >> then I I grew up as a galaxy person, uh, and I'm still fascinated by galaxies, and I well, it's really two things. One one is that we which is really just about the galaxies, which is that and this has been a mystery since I was a graduate student. >> Okay. >> Uh and it's still as far as I'm concerned still mysterious if we even if we have some

34:50better knowledge. Um which is that the most massive galaxies in the universe are not forming any stars. They're no stars forming in them. They are They are They are filled with old stars that mostly formed billions of years ago. >> Okay. >> And although you would think they would have available gas and some cases you see the available gas >> Mhm. >> uh which is normally what new stars form out of, they're not actually doing that.

Why massive galaxies stop forming stars

35:20Um and uh that problem, which existed in the '90s, is is still not, as far as I'm concerned, explained today. And that has consequences like those galaxies don't get as massive as your theoretical expectation would be. Now, there is a theory that is the that is sort of if there is a generally accepted theory, it would be this, which is that somehow the supermassive black holes that are accreting ma- matter at the centers of these very massive galaxies are preventing star formation from

36:02happening. >> Okay. >> And somehow >> Somehow. >> Somehow. And so so there are like you know at least half a dozen different numerical simulation methods where people incorporate this and successfully stop the stars from forming in these more most massive galaxies. All of the models work. >> Okay. they're all different. >> Okay, yeah, got it. >> So, they all start with there's a black hole at the center of the galaxy and and it's doing something and then they all do different things and then the end result is that they get the galaxies to have the right mass.

36:43>> Okay. >> Uh and but but the the the ways in which they do that in detail are all different. Okay. And uh and that to me is you know, I would like to know which of those answers is correct. >> Yeah. That's so interesting. Uh so, that seems counterintuitive. I mean, big galaxies should just have Yeah, cuz the Milky Way is forming stars, right? The Milky Way is not one of these. >> The Milky Way is probably is maybe 10 times less massive than >> Okay, than this size. Yeah. >> than than this happens, depending on exactly how you count. >> Um but uh but yeah, so the Milky Way is in this

37:24mass range where some some galaxies aren't forming stars, some are, and Milky Way's actually kind of in the middle. >> Mhm. >> Um but these guys it's like no one is forming stars in this >> It's very rare for for for them to be forming stars and and again you know, you need a lot of energy to stop the gas from flowing in and it's forming stars. You need to like heat it up or blow it out. You need a lot of energy. So, like people are like, these black holes, they can generate a lot of energy, which is a really good argument, but the exact way in which that happens is >> still a mystery. >> is still a mystery.

Dark matter and galaxy physics

38:00>> That's cool. >> So, that that that's one thing and I would say the other thing really for me, which is also related to galaxies is that I really think dark matter is a gigantic >> Yeah, of course. Everybody is >> And and I I I I one of the things that motivates me about studying galaxies, although they're very complicated and annoying in the way that that that it's hard to make predictions about them. Uh I do feel like if we really want to know more about dark matter, we better understand the physics of galaxies a lot better than we do today. Otherwise, I don't think we're making progress um without without sort of understanding

38:40the way the regular material works. >> Yeah. Yeah. All right. Totally. That's That's a fascinating problem. Um well, all the best to you and Carnegie, there's one last question that I want to ask. I want to go back down memory lane to your undergrad days. >> Oh boy. >> In at Cornell >> Yes. >> you did engineering physics, right? >> Yeah. >> Um but you actually were in the research group where they were doing numerical relativity simulations. >> Yeah, that's right. >> Right? Um so, have you seen Interstellar

Interstellar and black hole visualization

39:11the movie? >> I have. I have. >> And what did you think about those visualizations? Cuz I imagine that's kind of you know, the super you you were looking at numerical simulations around >> Yeah, actually Yeah, we were >> What did you think about that? >> We were doing visualizations. Uh >> Right. But with a much smaller computer, I bet. >> It was a much smaller computer. We did them uh we would We in fact we had to uh we did most of our work on green terms. Like these You know what these are? You know what I'm talking about? >> I I don't know actually. >> It's just You've got a keyboard. It's got a monitor that's 24 by 80 characters. >> Oh my gosh. Okay. >> Yeah, yeah. It's like you know, like in WarGames or whatever. Like it was one of those and then you know, on there Yeah.

39:51And then you were making these images and then to see the images, you walk over to a room where the monitor was. Anyway, it was a while ago. And then you wanted to put that video and then we would make videos for our professors to show at conferences. >> Nice. >> And like there was a whole 'nother room where it was like you slowly frame by frame put this onto a VHS. >> Yeah. >> Anyway, um it So, it was So, the visualization has gotten better. >> Yeah, definitely. >> Um, so what Interstellar does, I believe, um, and and which, uh, is is a really neat thing, is showing you like how the light actually

40:32how you would see it if you were in that space-time. >> Yeah, like and so they come >> Yeah, like do all the ray tracing. Yeah. So, so we did do ray tracing like similar to that. We were doing it, but we didn't do that for our visualization. We were doing that cuz there was a scientific question we were asking about Basically, we're asking like if you take two things, well, you know, we had these simulations that the graduate students and post docs had rhyme like of matter collapses forms a black hole. At that time, you couldn't make them do this. >> Okay. >> They could only come straight together. Uh, and they would form a black hole, but it's like when does the black hole

41:12form? >> Right. >> You only know the the definition of a black hole is the event horizon is when light can't get out. >> Right. >> No matter what. Like it forever can't get out. And and if you think about it, that's not really a local thing. You can't just say "Oh in this frame of this simulation at this particular moment, there's a black hole anywhere." You have to figure out if I shine light here, can it eventually get out? And if if if this if the thing's in motion, if the stuff is still flowing, then it's not obvious. >> Right, okay. >> Right? So, so so >> Which frame are you talking about and all >> Well, yeah, it's like well, it sort of

41:52depends on what these two clumps of matter do. If they're moving away from each other, maybe the light's fine. If they're coming together, there will be a black hole there, and this light won't get out. So, so we were doing ray tracing to figure out the answer to that question, like where the black hole actually formed. >> Right. >> Um, that was our job in the in the undergraduate group, but it was way like we would we could do a few thousand traces through the space-time. We we did not have the computing to like render an image rather through ray tracing. That that was that was beyond our capability. So, the stuff in Interstellar is is

42:32amazing in that way where these days you can you can literally make a picture of of what what you would see in in these situations. >> Yeah, and I think Kip Thorne's group actually did publish a few scientific results through the you know, the realization process of Interstellar, which I think is fascinating. >> Yeah. >> Um, well, it's been a great conversation with you. Thank you so much for taking the time. >> Um we really look forward to what you're going to do here at Carnegie. >> So do I. >> Yeah, and um, yeah, we'll we'll we'll see we'll see what happens. Thank you. >> Great. Thanks. >> Yeah. >> [music]

43:20[music]

Closing thoughts

43:26[bell]