Planetary Radio • Apr 08, 2020

The Slime Mold and the Universe

On This Episode

Joe burchett

Joseph Burchett

Observational Astronomer for UC Santa Cruz

Oskar elek 350

Oskar Elek

Computational Media Researcher in the UC Santa Cruz Creative Coding Lab

Betts bruce headshot 9980 print

Bruce Betts

Chief Scientist / LightSail Program Manager for The Planetary Society

Kaplan mat headshot 0114a print

Mat Kaplan

Planetary Radio Host and Producer for The Planetary Society

How could a lowly slime mold help researchers understand the distribution of dark matter and galaxies across the cosmos? Joseph Burchett and Oskar Elek of the University of California Santa Cruz will tell us about their team’s groundbreaking work. Bruce Betts and Mat Kaplan announce the first live and interactive What's Up segment is coming on April 23rd. The guys provide their usual assortment of space oddities in this week’s regular segment.

Galaxy cluster IDCS J1426
Galaxy cluster IDCS J1426 Galaxy cluster IDCS J1426 is located 10 billion light-years from Earth and weighs almost 500 trillion suns. NASA / ESA / M. Brodwin (University of Missouri)

Related Links

Trivia Contest

This week's prizes:

Bruce and Mat will record an outgoing message for your phone, if you dare.

This week's question:

Which X-15 pilots later flew on NASA space missions?

To submit your answer:

Complete the contest entry form at https://www.planetary.org/radiocontest or write to us at planetaryradio@planetary.org no later than Wednesday, April 15th at 8am Pacific Time. Be sure to include your name and mailing address.

Last week's question:

What mission saw the first musical instruments played in space? (Human performance on a real instrument or instruments. Live. In space.)

Winner:

The winner will be revealed next week.

Question from the March 25 space trivia contest:

Who was the first person to do a deep space EVA (extravehicular activity or “spacewalk”)? Deep space is defined as beyond low Earth orbit.

Answer:

The late Al Worden, Command Module Pilot for Apollo 15, was the first human to do a deep space EVA (extravehicular activity or space walk) when he retrieved film cassettes from the service module.

Transcript

Mat Kaplan: [00:00:00] The slime mold, and the universe, this week on Planetary Radio. Welcome, I'm Mat Kaplan of the Planetary Society with more of the human adventure across our solar system and beyond. Our main story this week may blow you away. Actually, it should. We'll talk with two scientists who have used the feeding behavior of one of our planet's lowliest creatures to help model the distribution of dark matter and galaxies across the cosmos. It's a tribute to the benefits of multidisciplinary science. Bruce Betts will also be along shortly with What's Up. Bruce and I have a big announcement too. Here's a preview.

You will be able to join us for the very first What's Up Live coming soon to a device near you. We don't have all the details yet but it will definitely be on Thursday, April 23rd at 1:00 [00:01:00] PM Pacific, 4:00 PM Eastern, or 2000 hrs, 8:00 PM UT. Just a little casual gathering as we all shelter in place and flatten that damn curve. And if you hear this in time, you can also be part of the very first virtual Yuri's Night celebration on Saturday, April 11th. Yeah, the world space party is also going digital this year. You can learn more at yurisnight.net.

Lastly, my terrific live conversation with NASA Chief Scientist, Jim Green, and astrobiologist, Penny Boston about life on Mars is now available on demand. The easiest way to find it is from exploremars.org. Here's an even quicker than usual review of headlines from the Downlink. The Planetary Society's weekly gift of great resources to fuel your love of space. NASA has begun a month long celebration of the Hubble Space telescope's 30th birthday. You [00:02:00] can even find images it captured on your birthday. CHEOPS is ready to start revealing the diameters of worlds orbiting other stars. That's the European Space Agency's characterizing exoplanets satellite. And NASA has selected SpaceX for the delivery of cargo to the gateway. That small maneuverable space station the agency plans to put in lunar orbit. As always, you'll find all the cool stuff in the Downlink at planetary.org/downlink, and you can also sign up to have it delivered every Friday for free to your inbox.

Understanding the distribution of dark matter across the universe is one of the greatest challenges faced by astrophysicists and astronomers. Surpassed, perhaps by learning what the crazy stuff is in the first place. We can't see it, but its gravity appears to give shape and substance to the largest collections of regular matter in the cosmos' galactic [00:03:00] clusters. Many theories and complex models have attempted to unravel what looks like a rat's nest of feathery filaments. We can probably assume that none of these efforts mean a thing to physarum polycephalum. You may have seen it growing across decaying logs or leaves in a forest or even on your front lawn. The spongy bright yellow mass is just a slime mold.

Nevertheless, the efficient way in which it seeks food has inspired a team of scientists. They have created an algorithm that may do a better job of modeling those filaments of dark matter than any previous attempt. The team includes scientists from the University of Massachusetts, North Carolina State University, and the Pontifical Catholic University of Valparaiso, Chile. Along with several researchers at the University of California, Santa Cruz. That's where the two lead authors of the March 10 paper are. Joseph Burchett is a UCSC astronomer. [00:04:00] His colleague comes from a discipline that may surprise you. Oscar Elek is a computational media researcher in the university's Creative Coding lab. He has a background in computer animation.

How these two young scientists came to realize that a slime mold can help us understand the universe is just part of what we covered in a recent conversation. Joe, Oscar, welcome to Planetary Radio. Congratulations on this work, which as soon as I got the press release, actually three press releases, I was just blown away. I am so glad to get you on the show to talk about this tie-in between a lowly slime mold and the largest structures in the universe. Welcome to the show.

Joe Burchett: Thank you happy to be here.

Oscar Elek: Thank you, Mat.

Mat Kaplan: As Jeff Goldblum told us in Jurassic Park, life finds a way. I don't think he had this in mind, but a, are you as amazed as I am by the fact that, that this living creature was able to help us learn [00:05:00] something about, I already said it, the biggest structures in the universe, Joe?

Joe Burchett: Absolutely. Uh, you know, this is trying to characterize, map out the cosmic web in the universe on, on the largest scales, uh, using only the galaxies that we can go out and observe that, uh, should trace this cosmic web. This has been a difficult problem that many people have approached in a lot of different ways. Each of these methods that people have employed to try to address this problem, um, have their strengths and, and weaknesses. It's taking a lot of effort, and, um, a lot of people, a lot of really smart, talented people have lent their energies toward this problem.

And when we started out to try and, and study the cosmic web, for our own purposes, I very much had in mind, um, all of this work that had, had sort of come before, and was about to default to, to any of these, these other established, uh, more conventional methods. Uh, [00:06:00] given that all of this effort had, had come before, it was especially surprising to me that turning towards, uh, this algorithm inspired by the slime mold, uh, was, was really, uh, the best application for our purposes.

Mat Kaplan: Joe, I read that you treated with some skepticism when Oscar told you that... Uh, and Oscar, I think I'm getting this right that you suspected that the way this slime mold grows, the way it spreads across the surface looking for food might be helpful with this challenge.

Oscar Elek: Yeah. I think initially the, the similarity was kind of visual. So, my, my history and my background is in computer graphics, looking at how we humans look at things. How we see things, and how we can, uh, emulate it. And then lo and behold here we have this organism that looks very suspiciously like the cosmic web. Now, the organism is flat. You know, it grows in, in 2D. [inaudible 00:06:59] I [00:07:00] mean surfaces of things, you know, dead trees, rocks, dirt, whatever.

Mat Kaplan: Yeah, flatland.

Oscar Elek: Flatland. Exactly. And the cosmic web is, you know, fully beautifully 3D webbed network. But it's kind of like the characteristic features that stood out. You know, when you actually look at it closer, and that's when we had already had the hindsight of having done, you know, some months of work on, on this project. We saw this, this property of the transport network arise that we realized is kind of the main link between these two.

Joe Burchett: Oscar mentioned that. You know, the, the initial connection that he made was, was really a visual one that the, the networks that the, the slime mold produces very much resemble, uh, the, the filamentary structure of the cosmic web. Now, our initial sort of intent was to find a way to, to visualize the putative cosmic web, uh, given our galaxy sample that we were We're working with for this project. I think even at, at the [00:08:00] initial phase when, when we first... Uh, when, when Oscar, uh, you know, brought this slime mold methodology to the table. At that phase, we were still thinking, uh, in terms of visualization. You know, it was even another layer of surprise to me that, that methodology could then carry through to actually doing quantitative analysis on the, on the data.

Mat Kaplan: Mm-hmm [affirmative]. I want you to introduce us to this, uh, this slime mold star, and I read, in fact, I saw the picture of it. And we will direct people from the episode page to, uh, this excellent UC Santa Cruz press release about this story. About this discovery by Tim Stevens up there at UC Santa Cruz. But first of all, I'm a product of the UC so I know the UC Santa Cruz mascot, which happens to be a [crosstalk 00:08:53]. Yeah, tell me you didn't choose this mold-

Oscar Elek: [laughs].

Mat Kaplan: ... because the similarly colored UCSC [00:09:00] mascot, the Banana Slug.

Oscar Elek: I really did not choose it for that reason.

Mat Kaplan: [laughs].

Oscar Elek: I did not actually. It is a complete coincidence.

Joe Burchett: [laughs].

Mat Kaplan: [laughs]. Anyway, if anybody hasn't seen a Banana Slug, maybe we'll link to, to that as well.

Oscar Elek: [laughs].

Mat Kaplan: It, it, it is... I'm a product of UC Irvine. We thought we were pretty cool when we chose the Anteater as our mascot, but I'll tell you the Banana Slug has us beat, uh, hands down.

Joe Burchett: [laughs].

Mat Kaplan: Um, tell us more about this, this slime mold and how it makes its way, makes its life on Earth.

Oscar Elek: Yeah, Joe. Joe, go ahead, and, and then I'll provide mine.

Joe Burchett: Just to set the stage for all of that inspired by this, this sophisticated, almost intelligent behavior, the slime mold exhibits. You know, we've, we've leveraged that to our particular astrophysical application. Uh, the slime mold is not an animal. It's not even a plant when we're thinking about the kingdoms, right? Of, of organisms, uh, on Earth. It's not an animal. It's not a plant. It's not even a fungus, but it's a [00:10:00] Protist. So, you know, in terms of the hierarchical complexity of, of life, it's a very sort of simple unicellular organism.

Mat Kaplan: It's, it's way down there, right?

Joe Burchett: [laughs]. Yeah, right. Right. So, this, uh, particular organism that the physarum polycephalum, um, has a phase in its lifecycle where the organism can essentially, uh, flourish. When, when it's flourishing, it's, and, and, and food is, is sort of abundant, um, say, in the damp dark forest floor where bacteria are growing. And so, food sources may exist in sort of desperate places, but the slime mold is able to establish a network. A unicellular macroscopic organism was able to establish this network to sort of connect from food source to food source in say across the surface of a log, and just across the ground, um, [00:11:00] moving from place to place. And it's really sort of this phase of its life, uh, where it's active. Where food is abundant, and, um, it can actually feed is, is kind of the phase of its life where it's inspiration brings about. It has sort of a filamentary structure. And in, in turn, this filamentary structure is what, uh, we use, uh, to map the cosmic web.

Oscar Elek: I, I look at the slime mold from, uh, from the position of just the patterns that it creates. Joe set the stage for the natural behavior of this thing. What it does from computational perspective, it actually finds or it actually solves an optimization problem. So, how do I relocate parts of my body to most efficiently cover as much space as possible?

Mat Kaplan: Mm-hmm [affirmative].

Oscar Elek: This is what actually makes the, the, the transport network sort of optimal. Of course, there is nothing like perfect in nature. It's just that it's very close to what you will build if you want to, to most efficiently connect a [00:12:00] bunch of spots on the map or a bunch of places in some higher dimensional space like galaxies. And so, it's this structural or spatial intelligence of the organism that, which we appreciate it for. And that actually helped us solve this problem. There was a bunch of abstraction necessary to happen for that. But in essence, we were still connected to the original behavior.

Mat Kaplan: You, you're not actually using the growth of the actual slime mold. You've built its growth, the intelligence [laughs]. If we can call it that, this, this what appears to be intelligent behavior, but probably not intelligence the way we humans think of it into the algorithm, which actually is the model that you've been able to use for this, I know. And I know-

Oscar Elek: Right.

Mat Kaplan: ... also, you weren't the first to notice that the slime mold, this network of filaments that it puts out might provide useful data and, and solutions, approaches for, [00:13:00] for seemingly unrelated problems. I mean, I read something about the Tokyo train network that, uh, actually also seems to resemble this.

Oscar Elek: Yeah. Um, this, this is a perfect example. Because here you have this, this batch of cities or districts of a city, and you need to build efficient railway. Now, the railway has been built al- already. But the, uh, work that you're referring to is just showing that what are the similarities between the organism and the real world problems that you're solving? And again, we are looking at this not from the perspective of they are the same rules that guides building these networks. That's not the case. It's more like what is the overall, um, what are the overall properties or behavior, uh, after the fact. And then what can we find they have in common that we can use to, to kind of find connections computationally?

Mat Kaplan: Joe, at first glance, you hear this, and it just seems nuts. I mean, a slime [00:14:00] mold's growth is not driven by the forces that form and arrange galaxies, right? I mean, that's kind of what Oscar is saying.

Joe Burchett: Yeah. Hence my skepticism at the very beginning. [crosstalk 00:14:10].

Mat Kaplan: [laughs].

Oscar Elek: [laughs]. A non believer.

Mat Kaplan: [laughs]. Well, a true believer now.

Joe Burchett: Oh, yeah.

Mat Kaplan: Didn't art play a part in this discovery?

Oscar Elek: Oh, yeah. The, uh, original connection was actually completely random. So, I heard about slime mold before. I'm pretty sure Joe did before, but you know, we just didn't connect the dots. And then we saw work of these, uh, of these media artist, [inaudible 00:14:36], who's based in Berlin. This guy just created this beautiful animations just using the base algorithm that we started with, uh, the work of Jeff Jones. He, he made it so beautifully stylized, even if just in black and white. The point is that you don't know what's, what's underneath it on the first glance. But the behavior the, the, the patterns and [00:15:00] structures that it creates are just irresistible. And then when you look at the cosmic web, which I had the chance to look at just a month before because we discussed this problem with Joe. Things just fell into, into place. So, yeah, this, this has been very important.

Mat Kaplan: Joe Burchett and Oscar Elek of UC Santa Cruz will tell us more about the slime mold and the universe in not much more than a minute. Planetary Radio is once again brought to you by the Caldwell Vineyard, creator of Rocket Science wine. Sadly, we've finished our bottle of this great proprietary red, but the fun hasn't stopped. I've somewhat foolishly told a spacey friend of mine that I'll give him the distinctive empty with its more than fun back label. The entertaining and thought provoking text on that little sticker was written by someone who calls herself [Kungsville Ava 00:15:51]. She was awarded a case of Rocket Science for her trouble.

Think you're up to the same challenge? Caldwell wants your most [00:16:00] creative work for their latest label writing competition. You've got till April 28th, and up to 80 words to share the inspiration and enjoyment Rocket Science offers anyone who pours themselves a glass. Enter as often as you like. Become a finalist, and you too will be enjoying a case of Rocket Science displaying your own work. Get the lowdown at caldwellvineyard.com/rocketscience, C-A-L-D-W-E-L-L. That's caldwellvineyard.com/rocketscience. Good luck, and happy landings. Back now to our conversation with Joe Burchett and Oscar Elek.

Joe, how did you begin to test this hypothesis? I, I, I know it involved this huge data set of, of galaxies.

Joe Burchett: The original project this is a Hubble Space Telescope archival data, uh, research program. This program is funded by the Space Telescope Science Institute, [00:17:00] uh, to essentially look at the large scale distribution of galaxies. Um, which large scale distribution of galaxies forms the cosmic web, right? Uh, and to use the, uh, the background quasars that shine through the, the cosmic web in the foreground. By studying the light of those background quasars, we were able to study the gas that comprises the cosmic web, um, or that, that is, um, a constituent of the cosmic web via its imprint on the background light.

Uh, but essentially, um, the key data sets here. There, there's sort of two, two main data sets. One is from the Hubble Space Telescope, uh, as I mentioned, but another is a sample of galaxies, um, that were taken from the Sloan Digital Sky Survey. This is enormous, an enormous survey. Millions of galaxies that are arranged over, you know, about a quarter of the night sky. I had selected a small sample of these galaxies [00:18:00] because if you essentially plot up, um, and this is true of, of any galaxy survey. If you just sort of plot up the locations of the galaxies, one can sort of intuit where the cosmic web structure should should, should fall.

Galaxies are apparently arranged in these sort of filaments, and they're, um, seemingly empty spaces that are voids. This was sort of the initial application once Oscar went away for a weekend, and, uh, he and his friend holed up and had their own little hackathon [laughs]-

Mat Kaplan: [laughs].

Joe Burchett: ... to, to do a lot of the work on-

Oscar Elek: Good times.

Joe Burchett: [laughs].

Mat Kaplan: [laughs].

Joe Burchett: To do a little work on the... To do a lot of the work on this, uh, on, you know, this algorithm, and, and sort of the key, uh, innovations that, that they added to it. So, we initially fed it the sample of galaxies and the intuition, just what you see, as I said, when you plot up the, the, the locations of galaxies. You know, where you kind of intuit the, uh, the filaments should lie. [00:19:00] When Oscar employed his algorithm, uh, with this data set what visually emerged was just such, uh, a striking match to you know that intuition. This was, you know, really an aha moment for me and I, you know, erased 99% of my skepticism.

Mat Kaplan: [laughs].

Joe Burchett: [laughs]. But, but the scientist in me couldn't let go of, of all of it. Um, and so, uh, the next step was to actually try to use this model in a, a situation where we sort of knew the ground truth about where, uh, the filaments existed, and where, where the filaments formed, uh, in the universe. Uh, we employed a, a dark matter only cosmological simulation. So, what that means is, this is essentially, um, an effort to put in the rules of cosmology as we know them, including the fact that the universe is dominated by dark matter, the universe expands. Using those just from the sort of some initial [00:20:00] conditions, the rules of cosmology, let the universe evolve inside a computer until the, the, the times of today. And then one can study how structure forms, um, under the influence of, of gravity and dark matter.

And so, it's in those simulations where, you know, we, we know that the cosmological theory generically predicts from simulation to simulation to simulation generically predicts this cosmic web structure. So, we used, uh, one of these simulations, much of the effort in developing the particular one we used, uh, was also to place here at UC Santa Cruz, taking the locations of where galaxies are supposed to form in the simulation, we ran the slime mold inspired algorithm in those locations. And that gave us a prediction of where the cosmic web filaments should be according to, according to the slime mold, right? But the nice thing about using this dark matter simulation is unlike in the real universe where we don't know exactly where the dark [00:21:00] matter is, in the simulation you know where the dark matter is.

And so, we could compare it to a one to one comparison of the slime mold prediction, uh, of where dark matter filaments should be, and where they actually were in the cosmological simulation. And we found an extremely tight correlation, uh, particularly at in the density regime where this project, um, that I described before this, this Hubble space telescope archival, uh, program, where, you know, we were really focusing our efforts and trying to learn more about, uh, about the, the gaseous ecosystems, uh, of the cosmic web and, and where galaxies live.

Mat Kaplan: So, it matched up pretty well, Oscar.

Oscar Elek: Yeah, this matched. Um, I just wanted to emphasize that it's not because the slime mold somehow magically is the same thing as the, as the, you know, cosmic web in the dark matter simulation.

Mat Kaplan: Mm-hmm [affirmative].

Oscar Elek: Like it is by design that we managed to find this match. What, which I'm saying just to not make the impression that, [00:22:00] that this is some kind of black magic-

Mat Kaplan: [laughs].

Oscar Elek: ... is really that the algorithm that, that we created is fitting to this data. This is a pretty standard, you know, procedure in machine learning and optimization where you have a data set. You, you're trying to, uh, approximate it or, or, um, estimate it with some kind of, uh, model, right? And in this case, the model is the obstructed behavior of the slime mold. The fact that we have this, this match is because we bent and configured and fiddled with the model until we got this, this really good match, and is really just that it kind of learns the structure.

Mat Kaplan: So, no need for a, a mystical connection here between slime molds and galactic, uh, clusters?

Oscar Elek: Yeah, let's leave that down till the end when we talk about the implications of this. Um, this is just computer science. [crosstalk 00:22:50].

Mat Kaplan: Uh, I hope that... [laughs]. So, far. I hope that people will visit this, uh, press release. I, I said, we will link to it from this week's show page, [00:23:00] uh, at planetary.org/radio because you will see, uh, all these great graphics, uh, that show the result of this computer modeling, and show you where the galaxies are, and then show you where the filaments are that, that connect them. And do I understand correctly, Joe? It's sort of at the intersection of these filaments where we tend to see galaxies forming.

Joe Burchett: Yeah. In particular, the sort of largest structures, the most massive collections of galaxies really form at the intersections of these filaments, the so called, uh, nodes. So these are galaxy clusters and galaxies super clusters form at these nodes. Uh, galaxies actually form all through the filaments themselves as well. Albeit, at, at much lower densities. So, if you think about the intersections of filaments are, are really kind of, uh, kind of the, the megalopolis. Huge cities-

Mat Kaplan: [laughs]. Yeah.

Joe Burchett: ... of the universe.

Oscar Elek: [laughs].

Joe Burchett: And, and lots of interesting things happen, happen in the big city. [laughs]. Uh, [00:24:00] in terms of, of the [inaudible 00:24:01], the galaxies and the gas that, that, that live in them. A big part of this study was to, uh, to try to understand the connection between where galaxies, uh, live. Uh, whether they live in the New York cities and the, and the Tokyos, and, and the Beijings of the world, of the universe, or whether they live, uh, out in the boondocks, um, in say the Paintsville, Kentucky where, where I grew up-

Oscar Elek: [laughs].

Joe Burchett: [laughs]. In a tiny little town way out in the country. Galaxies that live in relative isolation seem to have a tendency to form new stars. They, they, in one view, you could think of those galaxies living sort of longer, healthier lives in terms of it seems like they, they typically have young stars in them, which means they've had recent star formation. Galaxies that live in the big city, uh, tend to be dead [00:25:00] in the same view. [laughs]. I mean, there's [crosstalk 00:25:02].

Oscar Elek: [crosstalk 00:25:02] a similarity to this, to the society.

Joe Burchett: Yeah, right.

Oscar Elek: [laughs].

Mat Kaplan: The, the, the me- metaphor works, yeah.

Joe Burchett: [Laughs]. Uh, but you know, these, these galaxies haven't formed stars in a very long time.

Mat Kaplan: Mm-hmm [affirmative].

Joe Burchett: Just trying to understand that process of why do galaxies out in the boondocks, um, form stars profusely, and those that are in the big city, what's happening? And we think that's related to the gas supply out of which galaxies form stars. That's sort of the, the impetus for, for, for thinking about galaxies from an ecosystem perspective, and thinking about the gaseous environments, and, and transfer that must take place.

Mat Kaplan: This formation of stars, I, I read that this was also another of well, you called it apparently another sanity check as you looked into this, because you looked for the formation of these, these newer stars, and, and it helped increase your confidence.

Joe Burchett: Yeah, absolutely. The picture that I just painted of, um, you know, galaxies in sparser [00:26:00] environments, uh, having a greater tendency to be star forming than those that live in denser environments. Uh, this has been observed for decades now. So, so that's sort of a, a well known result. And it's been, uh, refined, uh, through, through the decades. Through the years, uh, in terms of quantitatively, uh, assessing this relationship. Uh, so yeah, so one of the sanity checks was, you know, the, the great thing about the product of this slime mold algorithm and, and running it on the data is we get a local density value for every point in our, our 3D space, right? We can go to the locations of the galaxies, and we can from the slime mold model, get its density, right? The density of its local neighborhood.

Mat Kaplan: Mm-hmm [affirmative].

Joe Burchett: And then, uh, we're able to, uh, look at the star formation activity of each galaxy and simply correlate the two and we're able to recover this behavior that I was, uh, that I was just referring to. The tendency of galaxies in denser environments to have a fixed mass. If you go into a denser environment, [00:27:00] the tendency for that galaxy to be read and dead, so to speak, is, is indeed increased.

Mat Kaplan: I am no astrophysicist or cosmologists. I just make my... I make my living talking to people like you. But this would seem to indicate a fairly major advance in how we can learn more about our universe, about the cosmos, and where it is headed, and perhaps a bit more of information about this mysterious stuff called dark matter.

Joe Burchett: Yes, absolutely. This study has been just sort of associating the, the gas, uh, of the intergalactic medium on the largest scales with the cosmic web that's traced by galaxies. And that cosmic web structure is consistent with predictions from our dark matter dominated, you know, cosmology, in terms of the, the matter in the universe. But yeah, there's, there's great potential here, and, and this is what I'm most excited about is, is where we could go from here. The, the [00:28:00] ability of, of this model, um, of this methodology to trace out the cosmic web structure. We have some real advantages here that I think some of the, the previously, uh, developed methods perhaps lack. It has maybe some weaknesses relative those other, uh, two previous models as well. But we're able to take large galaxy surveys, and very efficiently, uh, the simulation, the slime mold simulation runs in a matter of minutes. Um, we're able to produce a prediction of the dark matter cosmic web structure in the universe from the galaxy data in that survey.

Oscar Elek: You know, Joe mentioned this, this critical property that we get actually a density, uh, of the, of the slime mold or of the cost estimate in 3D space. M- Maybe you know, members of the audience might be curious, like, how can we get a density, right? We have, if you see a slime mold growing it's, it's kind of binary. Either it's there or it's not there, [00:29:00] right?

Mat Kaplan: Mm-hmm [affirmative].

Oscar Elek: So, this property comes again from the algorithm, um, and the modifications that we had to do to it. Again, we are kind of one step away from the original context in, in the sense that we no longer simulate slime mold, but we take inspiration, uh, from its growth. And then simulate a continuous structure that actually has, has density. And it can have gradients in space. It can have just full on transitions, and this is actually what was beneficial for the, the application but it's also something that makes it more, uh, robust because you're not trying to make a distinction between something being there or something not being there, but now it's a matter of degree to which it is there.

Mat Kaplan: So, it sounds like the human element in this was as important as the contribution of the slime mold.

Joe Burchett: [laughs].

Oscar Elek: Well, I mean, it, it taught us a lot and we are humble and accepting that something so simple can [00:30:00] be, uh, better than us in, in these aspects. But, you know, what I'm talking about is just the design that we added into it to solve our particular problem.

Joe Burchett: To, to get back to, to, you know, what we can learn about dark matter. Oscar just described this, this ability of the, of the methodology to give you sort of a local density and space or you know, a probability of a filament being at this particular location. A lot of people are searching fervently for, for dark matter and, and signatures of dark matter. Uh, this method can provide sort of a, a signpost or can provide guidance as to where to look, right?

Mat Kaplan: Mm-hmm [affirmative].

Joe Burchett: You want to look at the very densest pockets of the universe to try to, uh, have the greatest likelihood of detecting a dark matter signal, right? If there's, if there's more stuff, um, over here rather than over there you're more likely to detect it if you, if you look in the denser pocket. I think there's, there's a lot of potential to employ this method in those kinds of [00:31:00] searches as well.

Mat Kaplan: Very exciting. Shifting gears very slightly. Both of you embrace interdisciplinary approaches to science. It seems to me that we would not be having this conversation today if you didn't. Um, um, is this key to all of this?

Joe Burchett: Oh, yeah. Certainly. I began interfacing with, uh, with the lab Oscar works in shortly after I arrived in Santa Cruz. I was, uh, I was out at an open mic night. I, I'm a guitar player, singer, songwriter, where I went out to share a few of my songs. And another act, uh, featured Angus Forbes on drums. He runs the Creative Coding lab. We were just sort of chatting after both of our, our sets. Um, and he mentioned that he was an assistant professor at UCSC, and, uh, he was this expert in data visualization. He probably didn't phrase it exactly that way. Uh, he's a pretty humble guy. [laughs].

Oscar Elek: [laughs].

Joe Burchett: But yeah, so he, [00:32:00] uh, he, he mentioned that he's, you know, very interested in immersive representations of data and extending how, uh, humans interact with computers. Uh, in terms of, of analyzing data. This dataset from, from, from the project that, that we're working on here of all of these galaxies from the Sloan Digital Sky Survey immediately came to mind. I've, I'm like, I think I've got the, the data set for you. You know, if you're interested in, in sort of dabbling in, in astronomy, astrophysics. So, we started working on this visualization, uh, application, which, which we eventually, uh, published, and I can provide the link to that as well, if you would like.

Mat Kaplan: Sure.

Joe Burchett: So, yeah. So, so this is just a web based, um, sort of 3D representation of this data set juxtaposed with the Hubble Space Telescope, Quasar observations that are the diagnostics. Uh, what we use to, to diagnose the, the gas that fills the cosmic web, right? So that sort of began this, this cross [00:33:00] talk, you know, between me over here in astrophysics and Angus, uh, over there in, um, in computational media. It was, it was inspiring, um, being around, um, this, this group of people coming from a completely different perspective, but who really had fresh approaches to thinking about data and fresh approaches even sort of philosophically as evidenced by, right, the inspiration of art [laughs].

Oscar Elek: [laughs].

Joe Burchett: And data visualization. So yeah, yeah, I think this, this whole collaboration really epitomizes the, you know, the power of, of when, when people from totally different perspectives, but you know, sort of complementary philosophies can collaborate and bring each other's strengths, but also sort of inspire each other in unique ways.

Mat Kaplan: Oscar before you get into your own interdisciplinary leadings there's one other thing I want to know because Joe mentioned that he was a professional musician. I know you've recorded albums, Joe, and that one of your bands [00:34:00] was called the Mandelbrot.

Joe Burchett: [laughs].

Mat Kaplan: Uh, and Oscar-

Oscar Elek: Oh, oh, I did not that.

Mat Kaplan: [laughs]. Oscar, you're kind of a, a fractal guy, aren't you?

Oscar Elek: Yeah, very much so, yeah. You just blew my mind. I, I didn't know that you had such a band.

Joe Burchett: [laughs].

Oscar Elek: Uh, I know he had bands, but this, this, he hid from me.

Mat Kaplan: He talked about finishing a set at this open mic night. I, you didn't know he was talking about the Mandelbrot set.

Joe Burchett: Yeah, well, I played the [crosstalk 00:34:26]. I played the Mandelbrot's biggest hits.

Mat Kaplan: [laughs].

Joe Burchett: [laughs].

Oscar Elek: I mean, the guitar even kinda looks like the Mandelbrot, so, you know, there might be something there.

Mat Kaplan: [laughs]. Maybe, maybe, sounds like another project to look into. [crosstalk 00:34:38].

Oscar Elek: [laughs]. Well, um, [inaudible 00:34:41]. Okay, check.

Mat Kaplan: [laughs].

Oscar Elek: So, yeah, it's honestly I've, you know, seen as the early days of my studies, I, I've kind of lived in the fractal universe in the sense that fractals are very essential structure in, in graphics. You know, they, they're used to generate, [00:35:00] uh, visual content like terrains, and, and textures of rock materials and just generally, kind of the paradigm to think about all this visual complexity that surrounds us. We people build, you know, very orderly structures, but where in nature do you get square houses?

Mat Kaplan: Mm-hmm [affirmative].

Oscar Elek: You know, or cubic houses or circular wheels. You know, this is extremely rare. So nature copes with things by building structures on multiple levels, and that's why we get complex ecosystems, right? That's why we get all these, this amazing stuff that's on this planet and surrounding it. So, this is just kind of the prelude, but to me thinking about the cosmic web as a, as a fractal was extremely intuitive because that's how I've been, uh, approaching things for at least a decade now. Coming to Angus' lab for the, the Creative Coding lab. The nature of, of the, this, this competition media and research is that you involve the arts. You, you [00:36:00] involve the, uh, visual, you know, the acrostic, the, the different modalities to think about things.

Oftentimes, it's not as, as rigorous to approach things like that. But you know, rigor is something that for me follows, uh, kind of an initial inspiration or inception. Uh, and in this case, the inspiration was just really this, this visual gut feeling when you look at these structures, and so, yeah, this, this has been really important.

Mat Kaplan: Speaking of the arts, I have to mention one other thing. When I looked at this intergalactic network of filaments, uh, in that press release. I don't know if either of you is a, a Trekkie, a Star Trek fan. The audience knows that I am. There was once if I remember correctly in a Star Trek Voyager episode, a map, a computer generated map of what was supposed to be the Borg transwarp network to carry the Borg very quickly [00:37:00] around the, uh, Milky Way galaxy. [crosstalk 00:37:02]. And it reminds me of what I saw in the model that you guys have built. Is, is that crazy?

Oscar Elek: No. Like if I was a Borg I would definitely build my network as an optimal transport network. [crosstalk 00:37:18].

Mat Kaplan: [laughs].

Joe Burchett: [laughs]. Let's assume that the, the Borg has a simulate, has assimilated the, the slime mold I'm sure.

Mat Kaplan: [laughs].

Oscar Elek: [crosstalk 00:37:29].

Joe Burchett: [laughs].

Mat Kaplan: Well, just remember in the words of Captain Picard, resistance is never futile, and, and neither is interdisciplinary research guys. There's a poor, uh, segue for you. But this has been-

Joe Burchett: Beautiful.

Mat Kaplan: This has been delightful. Thank you so much, not just for the conversation today but for bringing these interdisciplinary interests, uh, to light, and making, uh, them work for us as we attempt to understand all that surrounds us, uh, including the structure of the universe [00:38:00] itself is this very exciting stuff. And I, I look forward to, uh, hearing how things continue to develop.

Joe Burchett: It was a very pleasant conversation. Thank you so much, man.

Oscar Elek: Yeah, thank you, and we are looking forward to that as well, trust me.

Mat Kaplan: UC Santa Cruz researchers Joe Burchett and Oscar Elek. Lead authors of a March 10th 2020 paper in the Astrophysical Journal Letters titled, Revealing the Dark Threads of the Cosmic Web. We've got lots of great related links on this week's episode page at planetary.org/radio. I'll be right back with Bruce.

Bill Nye: Bill Nye the Planetary guy here. You've heard Matt deliver highlights from the Downlink. Our great Space News Digest. You told us you want more. Well, you've got it. The Downlink now includes cool space images, and fascinating facts about the cosmos that you can share with your friends and family. Best of all, you can have the Downlink delivered to your inbox each week for free. Planetary.org/connect is where to go to learn more and sign up. That's [00:39:00] planetary.org/connect for the Downlink.

Mat Kaplan: We've reached the time for What's Up on this edition of Planetary Radio, so I'm joined by the chief scientist of the Planetary Society. That's Bruce Betts, and, uh, I want to amplify on that announcement I made at the top of this week's show that we will join each other once again. Except, for the very first time really live. Not just Planetary Radio recorded live-

Bruce Betts: [laughs].

Mat Kaplan: ... but live, live. [laughs]. Live and in person, virtually on the Thursday the 23rd. Thursday, April 23 1:00 PM Pacific, 4:00 PM Eastern, 2000, 8:00 PM UT as you informed me because you think in UT I think nowadays.

Bruce Betts: I do. [crosstalk 00:39:45].

Mat Kaplan: [laughs]. Well, that's good for us because I didn't have to look it up. I, I guess we're gonna call it What's Up Live?

Bruce Betts: Sure. That's catchy.

Mat Kaplan: [laughs]. Well, it's either that or Random Space Fact [00:40:00] Live, which is where we started with this. But it'll be fun because you can interact with us, uh, directly, at least through a chat. Ask us questions, make comments, submit a poem. Bruce will have all kinds of cool stuff for us, uh, to, uh, to talk about. I'll just be along for the ride. [laughs].

Bruce Betts: I will be the key personality addition to this show. Able to mock me when I can't answer a question.

Mat Kaplan: Yeah, as I so frequently do. Uh, anyway, that, there'll be more about this, uh, next week. And of course, we'll have more details as well at planetary.org, and I'm sure it'll be pushed out through all the Planetary Society's social channels until they realize what they've gotten themselves into at least. [laughs].

Bruce Betts: We just need to at least do the show before that happens. So, tune in [crosstalk 00:40:44]-

Mat Kaplan: That's right.

Bruce Betts: It'll be, it'll be o- online. It'll be virtual. It'll be a video and audio, and you'll be able to submit, uh, questions and comments. There'll be random space facts. There'll probably be trivia. It, it'll be all the joy that we give you every week, but live and, and [00:41:00] mockable.

Mat Kaplan: And longer. [laughs].

Bruce Betts: Yeah. Well, we're shooting for about a half an hour?

Mat Kaplan: Yeah, I think that's what we're shooting for. That, that, we're going to try and do lots of these live events out of the society, long overdue. We're going to do them at least weekly, and Bruce and I get to kick off this series.

Bruce Betts: Thursday, Thursday, Thursday. April 23rd.

Mat Kaplan: Be there.

Bruce Betts: Oh, yeah.

Mat Kaplan: [laughs].

Bruce Betts: We were going straight to the sky. In the, uh, morning east we've got those three planets all lined up, uh, from upper right to lower left. We've got super bright Jupiter, and yellow Saturn, and then reddish Mars. Mars and Saturn similar in brightness, and Mars will be brightening, and brightening, and brightening through October at its, uh, opposition. And if you check it out on April 15th, the moon will join the planets as well. Venus is still super bright in the evening West, and on April 26th the moon will join Venus. Venus is [00:42:00] kind of craning upwards in the sky relative to Orion or is Orion craning down. In any case, it'll be above Orion in the sky in the coming week. So, Comet ATLAS we talked about last week could be a naked eye comet in May or there are some hints that it's starting to break up so it may not be. So, as always with comets, we'll see.

Onto this week in space history, 1961, first human in space Yuri Gagarin. In 1970, Apollo 13 launched for what turned out to be an exciting trip. And in 1981, the first launch of the space shuttle.

Mat Kaplan: Big virtual celebration, uh, virtual Yuri's Night since we can't get together in person this time. It's, uh, it's coming up on, well for at least the main one. The consolidated one, I believe is Saturday, April 11 if you hear this in time. I'm sure it'll all be captured on video so that you can enjoy it after the fact, and, uh, [00:43:00] celebrate that passage of humanity in, into space just as Max the dog is.

Bruce Betts: [laughs]. That was actually Gracie.

Mat Kaplan: Oh, sorry Gracie. Good night, Gracie.

Bruce Betts: [laughs]. We move on to random space fact.

Mat Kaplan: I guess they aren't impressed by your, your impression.

Bruce Betts: Oh, they're not. Well, it's hard to bark words. That's why dogs don't talk, typically.

Mat Kaplan: [laughs].

Bruce Betts: So, Mat, speaking of isolation, Apollo astronauts to the moon were about 400,000 kilometers from everyone but themselves. Command module pilots who circled the moon alone where at times more than 3550 kilometers away from any other human. The equivalent of being alone in LA and having the closest human being Washington DC.

Mat Kaplan: Mm-hmm [affirmative]. Quite a distinction, and I think I got the gentleman who, um, represents the answer for the trivia question that you're about to, [00:44:00] uh, take care of for us, resolve.

Bruce Betts: It is not coincidental. All right. We move on to the trivia contest where we discuss this individual. I asked you who is the first person to do a deep space EVA? So, outside of, uh, low Earth orbit doing an extra vehicular activity, and, uh, I fear I once again have been caught not being specific enough, but it all worked out. How do we do?

Mat Kaplan: I am happy to announce this because Joe Murray, Joseph Murray in New Jersey has, uh, been a faithful entrant for about six years at least. He's finally won. Joe, you did it. And he says-

Bruce Betts: Yay.

Mat Kaplan: ... although, and he thought it was Dave Scott, but what we've read is, and, and as a lot of you have said that James Irwin sort of did a stand up EVA. Stood up in the hatch, but it was actually... The, the honors for the first real deep space spacewalk or EVA, extra vehicular activity go to his fellow crew person, Al Worden.

Bruce Betts: Yes, [00:45:00] indeed. Al Worden, and, uh, wanted to honor him. He just passed away, uh, two or three weeks ago.

Mat Kaplan: He was just a wonderful man. I, I will read what, uh, listen to Robert Laporte in Connecticut said. He said he had the honor of meeting Colonel Warden three times. Gentleman and a gentle person, special man in a very unique group of astronauts, those 24 brave men who went to the moon. All of them were men, of course. A terrific guy with a great sense of humor as well. Very lively, and, and I know this because I also got the chance to interview him for Planetary Radio, and, uh, we'll share that link once again. Uh, it, it's a very enjoyable conversation.

Bruce Betts: Yeah, that would be great.

Mat Kaplan: What Bruce was referring to about the confusion. A lot of you said Al Worden, but also hedged your bets by saying maybe Neil Armstrong because that, of course was a couple of years earlier with Apollo 11 on the moon, but that's not what you were looking for, right?

Bruce Betts: No, but I probably would have accepted either since [00:46:00] I was sloppy, sorry.

Mat Kaplan: [laughs].

Bruce Betts: But random.org chose someone who did say Al Worden, which was my vision of floating in space between the moon and the earth, and going outside your space capsule. That seems kind of wild.

Mat Kaplan: Mel Powell in California. He's one of those who hedged his bets, and he said, though, that... He says, "I, I'd say I'm starting to know how Bruce thinks, but that's just terrifying."

Bruce Betts: [laughs]. Save yourself.

Mat Kaplan: [laughs]. I thought you'd like that. Joseph, you are going to win... Well, the honor is not dubious, but the prize may be. If you choose, Bruce and I will record a message for you. A personalized message you can use it as your outgoing message on your voicemail system if you, if you'd like to do that or you can just, you know, maybe, uh, put it on your mp3 or Bluetooth alarm clock and wake up to us every morning.

Bruce Betts: [laughs]. Or put it on a loudspeaker broadcasting throughout the neighborhood.

Mat Kaplan: [00:47:00] The reason Al Worden went out was to get these filmed cassettes from the cameras in the service module of, uh, Apollo 15. They were big. As we heard from Ian Jackson in Germany. The biggest of these was 152 kilograms. Film cassette contained two kilometers of film 1,650 photos, which actually sounds kind of low. Ian says, "They don't make cameras like they used to." It was Mark [Dunnin 00:47:26] who said, "Yeah, what would it be now? A, a, little tiny wire on an SD card. He could just yank on the wire, I guess [laughs] and pull back into the, into the capsule." Well, they'd have Bluetooth, right? They'd have Bluetooth to the service module."

Bruce Betts: They certainly wouldn't have giant canisters of film that you had to leave. [crosstalk 00:47:43].

Mat Kaplan: No. Mark added, "Now more than ever you guys are the high point of my weak [my version of I love you man]."

Bruce Betts: Oh, man. We love you too.

Mat Kaplan: Finally, Jean [Lewin 00:47:55], a, our Poet Laureate has the, uh, has the week off. But Jean Lewin up in [00:48:00] Washington gave us this. Once a man took a stroll from earth so far away, opening endeavors door his scroll an EVA, retrieving a cassette of film. The reason for this peregrination, the first in deep space, one of three accomplished in this fashion. He holds a record that still exist in Guinness Books today. One for the most isolated human over 2000 miles away. Al Worden is the astronaut who took this distant trap. I would have mailed him a letter if I only had a stamp.

Bruce Betts: Oh.

Mat Kaplan: Nice work. Thank you, Jean. We're ready to go on.

Bruce Betts: We're headed to X15 pilots. I know you love the X15.

Mat Kaplan: Mm-hmm [affirmative]. I sure do.

Bruce Betts: Which X15... You may just know this, Mat, but don't say it. Which X15 pilots later flew on NASA's spacecraft missions? Go to planetary.org/radiocontest.

Mat Kaplan: Well, I sure know one of them, but I won't say who because you've asked me not to. Otherwise, I would [00:49:00] have you have. Uh, have [crosstalk 00:49:02]-

Bruce Betts: [laughs].

Mat Kaplan: ... this time. You have until the 15th. That'd be what would have been tax day here in the United States. But we all know that that's, uh, been swallowed up by the pandemic, uh, for, uh, and we get an extra three months. Wednesday, April 15 at 8:00 AM Pacific Time, and if you were chosen by random.org, and you've got the right answer, you also might get a little personalized message from Bruce and me that you can do with as you wish. That's it.

Bruce Betts: All right, everybody. Go out there. Look up the night sky, and think about what it would be like to be alone circling the moon. Thank you, and good night.

Mat Kaplan: Well, I can only tell you that Al Worden said he loved it. He loved every moment of it. He loved the solitude as well, and, and being further from any other human being than anybody else has, has ever been. And that's in the interview I did with him. Hey, I'll talk to you next week, and also, uh, on the 23rd when we will, uh, hear from everybody [00:50:00] else as well for whatever we will end up calling it. I'll say, uh, as a placeholder, What's Up Live for now. Thanks, Bruce.

Bruce Betts: Cheese muffin live.

Mat Kaplan: Well, do you have one in your hand? I bet.

Bruce Betts: No, I don't. But I'm going to.

Mat Kaplan: That's Bruce Betts, the chief scientist for the Planetary Society who joins us every week here for What's Up. Planetary Radio is produced by the Planetary Society in Pasadena, California, and is made possible by its never slimy members who want to understand the cosmos. Sound like you? Join them at planetary.org/membership. Mark Hilverda is our associate producer. Josh Doyle composed our theme, which is arranged and performed by Peter Schlosser. Be safe everyone, Ad Astra.