0:00

The The PC

0:02

gave us computing power at home, the

0:04

internet connected us, and mobile let us

0:06

do it pretty much anywhere. Now generative

0:08

AI lets us communicate with technology in

0:10

our own language using our own senses,

0:12

but figuring it all out when you're

0:15

living through it is a totally different

0:17

story. Welcome to leading the shift, a

0:19

new podcast for Microsoft Azure. I'm your

0:21

host, Susanet Linger. In each episode, leaders

0:23

will share what they're learning to help

0:25

you navigate all this change with confidence.

0:28

Please join and subscribe wherever you get

0:30

your podcastsasts Live your career further, faster

0:32

in just two days, with the

0:34

Harvard Professional and Executive Development

0:36

Program. Advance your leadership skills.

0:38

Craft smarter business strategies. Build

0:40

your network and transform how you

0:43

work and think to keep your

0:45

career moving forward. You'll earn

0:47

a certificate and can add

0:49

Harvard to your resume with

0:51

Harvard Professional and Executive Development. Learn

0:54

more at professional dot DCE, dot

0:56

Harvard, E-U, slash modify. Hey,

1:01

everybody, it's a great pleasure to

1:03

welcome you to the special

1:06

in -person episode of the Into

1:08

the Impossible podcast featuring two -time

1:10

guest, Constantine Batijan. Constantine's a renowned

1:12

astrophysicist, planetary scientist at Caltech,

1:14

a professor. It's the second time

1:16

on the podcast, and he

1:18

is one of my favorite guests

1:20

and especially infectious enthusiasm to

1:23

experience him in person. You'll hear

1:25

all about the quest to

1:27

uncover Planet 9. What the critics

1:29

are saying in their attempts

1:31

to assail and perhaps disprove Constantine, how

1:33

does he handle that? New research

1:35

that will be coming up involving

1:37

LST or the Virubin Observatory in

1:39

Chile, where the next frontier in planetary

1:41

research, not just in exoplanets, as

1:43

we've talked about way more than

1:45

this topic, this is inner planets,

1:47

planets in our own solar system

1:50

beyond the orbit of Neptune. And

1:52

last, but not least, you're going

1:54

to hear a deep dive into the

1:56

physics behind Jupiter's past history. Oh,

1:58

how can we do archaeology? on planets

2:00

in our own solar system. Well, Constantine's

2:02

figure out a way to do that.

2:04

And it involves, of all things, its

2:07

magnetic field. You'll hear about that, see

2:09

the latest research into the planetary dynamics

2:11

of our own solar system. Constantine's known

2:13

stranger to attention. You'll see him featured

2:15

in 60 minutes. Now, I want you

2:17

to sit back, relax, and enjoy this

2:20

episode into the Impossible with the irrepressible

2:22

Professor Constantine Battijian. Any

2:27

sufficiently advanced technology is indistinguishable from

2:29

magic. Open the pot bay doors,

2:31

Hal. Welcome back everybody to an

2:34

episode that promises to be out

2:36

of this world with a friend,

2:38

a two-time guest on the Into

2:40

the Impossible podcast, Professor Constantine Petitian.

2:42

Welcome, my friend. You survive fires

2:44

and floods and mudslide potential to

2:46

be here. Thank you so much.

2:48

We're not even talking about the

2:50

drink. That was just to get

2:52

down here. I mean, that was

2:54

a while. I always thought it

2:56

was a good day to come

2:58

down to San Diego whenever I'd

3:00

leave Caltech's confines and come down

3:02

here. I always felt it was

3:04

a good day. We're so happy

3:06

you're here. Good to be here,

3:08

man. The Hebrew planet, as I

3:10

call it, Jupiter. Jupiter. Yes. We'll

3:12

be talking about that. We'll learn

3:14

about how it got its size.

3:16

You collaborate with Fred Adams, I

3:18

think, on that. I sure do.

3:20

In paper, he's been a guest

3:22

a long time ago. I got

3:24

to get him back. And I'll

3:26

use this as an opportunity to

3:28

do that. But first we got

3:30

to talk about updates to the

3:32

most important topic, your band, how's

3:34

your band doing? Man, my band

3:36

is doing awesome. So we, you

3:38

know, we played a couple gigs,

3:40

I think, a tremendous amount of

3:42

fun. Like the club we play

3:44

at most often now, the mix

3:46

has this like huge screen behind

3:48

the stage. And so, you know,

3:50

we've been kind of incorporating that

3:52

into the show and also just

3:54

because like AI video production is

3:56

now so easy. It's been kind

3:58

of adding an extra element. We're

4:00

working on a new album. So

4:02

things are going, you know, knock

4:04

on wood, things are going well.

4:06

That's awesome. Okay, let's get to some

4:09

of the meat of the conversation because

4:11

we're gonna talk about lots of things

4:13

involving planets. I have some planetary swag,

4:16

you'll find out about that in just

4:18

a bit. But the first thing is.

4:20

Yeah, for someone who's not familiar, Planet

4:23

9. We used to have Planet 9.

4:25

Actually, this asteroid up here, you can

4:27

barely see it, it's called Aster, 66,

4:29

18, Jim Simons. Got that named after

4:32

Jim Simons. And it was discovered by,

4:34

see, the Discover, can you read that?

4:36

Clytombo. Yeah, now what's the importance of

4:39

Clytomba in the world of

4:41

planetary discoveries? Clytombo, famously discovered

4:43

Pluto. Of course, right, Pluto

4:45

was the original planet X.

4:47

Right, right, There was all

4:49

of this discussion about there

4:52

being an additional planet which

4:54

was largely driven by Lowell,

4:56

right? And like Lowell Observatory

4:58

was in part constructed in

5:00

order to look for this

5:02

elusive planet. And Lowell died

5:04

I think in 1916, but

5:06

the search kept going and

5:08

Clyde Tombo, who was employed

5:10

at Lowell Observatory in 1930,

5:12

discovers Pluto. One of the

5:14

things that I think many

5:16

people don't realize, when you

5:18

just discover something up in

5:20

the night sky, you don't know how

5:22

massive it is. And so you kind

5:24

of say, well, I was looking for

5:27

a thing that was supposed to be

5:29

seven earth masses, so it's probably seven

5:31

earth masses. But Clyde Tombo immediately realized

5:33

that, well, if it was something that

5:35

big, you should be able to resolve

5:37

the disk. And instead, it looked kind

5:39

of like a point source. So he was

5:41

like, probably one. Earth mass like you

5:44

know there's no way to calculate it

5:46

if it doesn't have a satellite and

5:48

you know you can watch Pluto's

5:50

mass kind of decrease throughout through

5:52

the literature and there's even some

5:54

joke paper from like the 80s

5:57

that makes a plot of Pluto's

5:59

mass at a function of time

6:01

between 1930 and like 1980 something

6:03

and predicts how Pluto would disappear.

6:05

It would cross zero in like

6:08

2005 or something like that. Anti-matter.

6:10

Yeah. And so, you know, it

6:12

was it was really realized only

6:15

in the 70s when when Sharon

6:17

the satellite of Pluto was discovered

6:20

just how minuscule the mass of

6:22

Pluto is. And so that was

6:24

the kind of story that led

6:27

to the demotion of Pluto, you

6:29

know, Mike, of course, my partner

6:31

in crime had a lot to do

6:33

with that back 20 years ago. But

6:35

the thing we're looking for now is

6:38

not Some minuscule thing right

6:40

the thing we're looking for is the

6:42

legitimate planet night And it would be

6:45

far beyond the orbit of even Pluto

6:47

correct about You know about factor of

6:49

10 15 further away. Wow nowadays we

6:52

don't use you know pencil and paper

6:54

like Lowell or even Laverier did right

6:56

I recently discussed with Davis Sobel who

6:59

wrote a lot of wonderful books including

7:01

a book called Galway of his daughters.

7:03

She talks about that And we were

7:06

just musing on how many, you know,

7:08

amazing discoveries Galileo made, but he also

7:10

discovered Neptune. He didn't realize he discovered it,

7:12

but he discovered it. So I aspire to

7:14

be like that. I aspire to be like

7:16

that. I aspire to my blunders, you know,

7:18

like Einstein's cosmological. It should be as good

7:20

as that. But, um. Nowadays we have end

7:22

body simulations and so forth and to me

7:25

as a scientist that opens up you know

7:27

a whole new realm including AI machine learning

7:29

and stuff but also potential pitfalls and I

7:31

wonder if you could respond you know some

7:33

of the critics say when you're talking about

7:35

this object which we call planet nine and

7:37

that you were at the very very forefront

7:39

of its of its investigation that

7:42

you know there could be artifacts

7:44

introduced because of these in-body simulations.

7:46

So can you explain why it's

7:48

so important to discover this? And

7:50

what are the new tools and

7:52

new pitfalls of those new tools?

7:54

Okay, so first of all, the

7:56

in-body simulation as a thing, right,

7:58

is in effect... a miracle experiment, right?

8:01

It is a realization of the

8:03

solar system as it unfolds over

8:05

its lifetime. You start off with

8:07

a reasonable initial condition and you

8:09

say, I've gotten to the point

8:11

where we are now at four

8:13

and a half billion years after

8:15

the formation of the sun, does

8:18

the solar system that I've created

8:20

in my numerical experiment look like

8:22

the one that we see? What

8:24

are the pitfalls? Well, the simplest

8:26

one is just in the method,

8:28

right? You can, if you're not...

8:30

careful, you can screw up and

8:32

you can introduce like fictitious dynamics

8:34

into your simulations, that's pretty easy

8:36

to get. Like at this point,

8:38

that's almost never a question, right? I

8:41

think much of the discussion, right, has

8:43

been, you know, related to planet nine,

8:45

has been, okay, you do this numerical

8:47

experiment, how do you then compare the

8:50

output, like the, what we see at

8:52

the end, to what you really see

8:54

on the night sky? And this is

8:56

where I think my collaboration with Mike

8:59

has been the thing that has, you

9:01

know, for a change made us greater

9:03

than the sum of the parts, not

9:05

less than the sum of the parts,

9:08

because, you know, Mike is an observational

9:10

astronomer. He's a ninja when

9:12

he comes to understanding,

9:14

right, what the night sky is telling

9:16

us. And we've been, I think, through

9:19

the kind of back and forth, which

9:21

sometimes gets kind of loud, but it's

9:23

all fun, you know, we've been able

9:25

to kind of challenge each other and

9:27

really get down to the question

9:29

of how do we take this output

9:32

and compare it meaningfully with what we

9:34

see on the night sky because when

9:36

you're observing stuff you don't just have

9:39

access to the entire solar system you

9:41

just have access to what you can

9:43

see. You know that leads to

9:45

a well-defined number which is

9:48

the false alarm probability of

9:50

this entire You know story and there

9:52

are different lines of evidence for planet

9:54

nine and like the two that I

9:56

think people like to talk about because

9:58

they're coming to eat to imagine is

10:00

that if you go far enough away,

10:02

all the orbits are all facing, they're

10:05

all like swinging out in the same

10:07

direction, and that has a well-defined false

10:09

alarm probability of about 0.2% right? There

10:11

are other lines of evidence like there

10:13

are somewhat higher, a signal like five-sigma

10:15

ones, so you can do all this

10:17

in a pretty rigorous way, but the

10:19

point I think is that you have,

10:21

you can't just... you know, do a

10:23

simulation and say, well, here's what I

10:25

got, you know, and, you know. So

10:27

when Galileo turned his telescope, here he

10:29

is there, not this actual guy, but

10:31

he turned his telescope to the skies

10:34

in 1609, he saw, I looked at

10:36

the moon, he saw it was flawed,

10:38

full of craters, mountains, he measured the

10:40

height of the mountains, he measured the

10:42

height of the mountains, the guy is

10:44

incredible. And speaking of the moon, this

10:46

is, you got to choose, which one

10:48

of these do you think is more

10:50

valuable actually? Yeah, well definitely a smaller

10:52

one. This one. This is a piece

10:54

of the moon. This was delivered not

10:56

by the NASA astronaut, this is Charlie

10:58

Vince, meteorite from your former. Chelyabins. Chelyabins.

11:00

Chelyabins. Chelyavins, Spassiba. This is a piece

11:02

of the moon. Okay, and the reason

11:05

I'm giving that to you is I'm

11:07

so grateful that you're here for your

11:09

second appearance and that is real and

11:11

you'll get also a piece of the

11:13

Proto solar system. This is a meteorite.

11:15

So this is a chunk of. Oh,

11:17

this is an iron meteorite. Yeah, this

11:19

is from the Campadisielo in Argentina, which

11:21

you will win too guaranteed if you're

11:23

out there and you have a.EDU email

11:25

address and you live in the United

11:27

States, Brankinginganking. give us to you is

11:29

because when Galileo discovered these four little

11:31

stars he called them after his funding

11:34

agency not the NSF but the the

11:36

Cosimo Medici family he called them after

11:38

them. Now we called them the Galilee.

11:40

What I'm getting at is that then

11:42

you know added four new moons to

11:44

the retinue of moons

11:46

that we knew

11:48

about in the solar

11:50

system alone. To

11:52

what extent, if we

11:54

found Planet 9,

11:56

would that essentially imply

11:58

Planet 10, Planet

12:00

11, Planet 12? Would

12:03

there be many,

12:05

many more to come,

12:07

essentially? Yeah, so

12:09

the solar system has

12:11

quite a bit

12:13

of real estate, right?

12:16

You can keep moving out. Eventually

12:18

though, you run out of

12:20

real estate that's stable, because eventually

12:22

you start to see the

12:24

galactic tide, right? And the galactic

12:26

tide, you know, functionally

12:28

basically just takes your

12:30

orbital inclination with respect to

12:32

the plane of the galaxy

12:34

and trades that for eccentricity

12:36

through something akin to what's

12:39

called the Kozai effect. And

12:41

in any case, you know,

12:43

there is sort of

12:45

half of an order of magnitude

12:47

left still in semi -major axis,

12:49

but once you go well beyond

12:51

that, passing stars, galactic tides start

12:53

to really mess you up. That's

12:56

why, you know, kind of where

12:58

Planet 9 is, or where we

13:00

infer it to be. So in

13:02

the region, we call the inner

13:04

Oort cloud, right? Where it's material

13:06

that could have been trapped there

13:08

by interacting with the solar system's

13:10

birth environment, like the cluster of

13:12

stars in which the sun formed.

13:15

Now that that's gone, there isn't really

13:17

a way to trap material there

13:19

anymore. And well beyond that, you're

13:21

in the Oort cloud. And the

13:23

fact that we have Oort cloud

13:25

comets, right, that come in is just

13:27

a manifestation of the fact that

13:29

if you're in the Oort cloud,

13:31

you're not just sitting there forever

13:33

orbiting, you know, happily, there are dynamics

13:35

that unfold. It will cause you

13:38

to be much more elliptical and

13:40

then eccentric and then eventually trans out

13:42

of the Kuiper belt, right? So

13:44

not saying it's impossible. I'm saying

13:46

it's into the impossible. You talk about

13:48

this thing, which I think is very

13:50

interesting, at least the names of it,

13:52

I love these names that you give,

13:54

the perihelion distribution dynamics. You talk about

13:56

this Planet 9 inclusive model being relatively

13:58

flat. This is some distribution - function in your

14:01

most recent paper that the graduate students will

14:03

be quizzed on later on today after your

14:05

wonderful talk. Talk about that. What is a

14:07

flat distribution to me? Oh, there's got to

14:10

be many of these things and and yet

14:12

I can understand because of your the clarity

14:14

of your presentation just now that actually may

14:16

not be the case that there's sort of

14:19

an infinite number of planets left to come.

14:21

I mean planets that we would say are

14:23

honest to goodness planets, not chunks of you

14:25

know, Pluto or asteroids. Yeah. So the most

14:28

recent paper is something that we were inspired

14:30

to do, and this is work

14:32

that, you know, I did with

14:35

Mike, but also with my close

14:37

collaborators, you know, Sandra Morbidelli, niece,

14:39

and also David Miss Worny at

14:42

Boulder. It's a Southwest Research

14:44

Institute. So, you know, for about

14:46

eight years or whatever, how long it's

14:48

been, I guess now nine years, that

14:50

we've been working on this planet

14:53

nine stick. We have always been

14:55

focusing on the most distant, the

14:57

most kind of untouched orbits possible

15:00

because Neptune messes stuff up. There's

15:02

a subset of the Kuiper belt,

15:04

which we kind of ignore because

15:06

we say, well, a close approach,

15:08

it hugs the orbit of Neptune

15:10

and the chaotic dynamics that

15:13

ensue from interacting with Neptune,

15:15

make that just as... Have you

15:17

ever felt lost in the vastness of

15:19

scientific knowledge? I'm Dr. Brian Keating,

15:21

astrophysicist and hosted the Into the

15:24

Impossible podcast, and I found a

15:26

way to navigate. The cosmic sea

15:28

of information overload. Enter Consensus. The

15:31

AI-powered search engine that's revolutionizing how

15:33

I do science and how we

15:35

all can explore scientific frontiers. Imagine

15:38

having a brilliant research assistant, available

15:40

24-7, capable of sifting through over

15:42

200 million papers in seconds. Now

15:45

I know many of you are

15:47

involved in academia from all the

15:49

feedback I get. And I know that

15:51

someone like you, who's always pushing the

15:53

boundaries of what's possible, you'll be skeptical

15:55

as I was at first. But consensus...

15:58

blew my mind. I describe it as

16:00

Google Scholar and Chatgy BT had

16:02

a love child with a PhD

16:04

in every scientific field, but without

16:06

the clunkiness of Google Scholar or

16:08

the hallucinations of Chatgy BT. Whether

16:10

I'm diving into the mysteries of

16:13

cosmic inflation or exploring quantum entanglement,

16:15

consensus doesn't just find papers. It

16:17

understands concepts, draws connections, and even

16:19

highlights scientific consensus with their novel

16:21

consensus meter. And here's the kicker.

16:24

I loved it so much I've

16:26

been paying for the pro version

16:28

long before they became a sponsor.

16:30

That's how game-changing it is. Ready

16:32

to supercharge your curiosity, finances fast,

16:35

and outwork the competition. Well, whether

16:37

you're a fellow scientist or someone

16:39

who... just craves asking, what if?

16:41

Consensus is your ticket to the

16:43

frontiers of human knowledge, all back

16:46

by science with references. I use

16:48

it almost every day to check

16:50

on things from my field, but

16:52

also from other fields, like the

16:54

field of medicine or energy, or

16:56

even controversial topics like climate change.

16:59

So try consensus now. Visit consensus.

17:01

and enter code Keating at checkout

17:03

for 40% off consensus premium for

17:05

two years. I wish I had

17:07

that before I started paying for

17:10

consensus, or visit the link in

17:12

the description below. Now back to

17:14

the episode. then the or cloud,

17:16

then the heliopause or something like

17:18

that. Whatever. Yeah. Right. Okay. So

17:21

you were saying, have to do

17:23

tugs on it. Right. So stuff

17:25

that is kind of in the

17:27

outer solar system whose evolution is

17:29

chaotic and stuff that's out, it's

17:32

kind of being thrown out of

17:34

the solar system as we speak.

17:36

So we kind of tend to

17:38

ignore that and only focus on

17:40

the orbits that are sufficiently detached,

17:42

sufficiently calm, that we know that

17:45

kind of gravitational gravitational. footprint of

17:47

planet nine has a chance of

17:49

being seen in these orbits. And

17:51

so what we thought about a

17:53

couple of years ago, I guess

17:56

a year and a half ago

17:58

now, was more like Well, let's

18:00

now look at the opposite extreme.

18:03

Let's now look at the

18:05

most unstable component of the

18:07

Kuiperbell. These are things that

18:09

orbit in the plane of

18:11

the solar system that physically

18:13

cross the orbit of Neptune.

18:15

And so they're being actively

18:17

scattered around. Now, if you

18:19

reason through a question of like,

18:21

should such objects exist in the

18:24

first place, the answer is basically

18:26

no, because their lifetime in

18:28

the solar system is 10. maybe a

18:30

hundred million years at most, so that

18:32

Neptune should clear the solar system out.

18:35

But if planet nine is there, then

18:37

planet nine should be systematically injecting these

18:39

things back into the solar system interior

18:41

to Neptune. And it would have to

18:44

clear out its orbit, right, to be

18:46

consistent with our friends of the only

18:48

union I'm a member of the International

18:50

Astronomical Union. Yeah. And so what we

18:53

did, yeah, so we did is we,

18:55

you know, conducted. numerical simulations

18:57

which I think are the most

18:59

kind of encompassing and body simulations

19:02

of the solar system of evolution that

19:04

you know have been done perhaps and

19:06

you know we asked the question of

19:08

looking at the this highly unstable component

19:10

of the of the kyper belt right

19:12

can we rule out rule in a solar

19:14

system with without planet nine and

19:16

what we found is that the

19:19

solar system without planet nine is

19:21

five stigma ruled out. And the

19:23

social system with planet nine is

19:25

indistinguishable from the data. So that, even

19:27

though it's kind of a surprising, you

19:29

know, it's surprising that the

19:32

most unstable kind of boring

19:34

part of the kitere belt

19:36

gives you the most statistically

19:38

significant thread, that's the most

19:40

stringent evidence we have that.

19:42

planet 9 is really out there. So

19:44

that's when you know we hear things like

19:46

yeah the five sigma confidence that's saying that

19:49

ruling out the null hypothesis that planet 9

19:51

does not exist. Very good very good. Now

19:53

how much of this you know depends on

19:55

on data when Laverier predicted the existence of

19:58

Neptune and then it was found the same

20:00

day or something like that the legend

20:02

goes are very soon after and then

20:04

of course he went on to blunder

20:06

and predict Vulcan right so you know

20:08

sometimes if you only have a hammer

20:10

you hit yourself in the head too

20:12

many times why is it so hard

20:14

I mean you know no no offense

20:17

but sure you told me I have

20:19

to look in this area of the

20:21

sky to see the CMB's B mode

20:23

polarization I'd be out there tomorrow with

20:25

the science observatory we'd be looking for

20:27

it would be looking for it. for

20:29

the Higgs boson to be awarded Nobel

20:31

Prize. So why is this so hard?

20:33

Yeah, well, I'm glad you bring up

20:35

the Laverier discovery of Neptune as a

20:38

kind of counterpart, because what

20:40

Laverier was able to calculate

20:42

very precisely was the acceleration

20:44

is coming from there. Okay,

20:46

and this had to be, this had

20:48

to do with the fact that

20:50

he was doing the calculation in

20:53

1846, and Uranus and Neptune happened

20:55

to be close to conjunction. And

20:57

so the. information that was stored

20:59

in the Iranian residuals was

21:02

actually not the mass of Neptune,

21:04

not the orbit, but like, where is

21:06

it on the night sky? We're in

21:08

precisely the opposite regime. What we can

21:11

calculate from the orbits, like the orbital

21:13

distribution of the kyper belt, is the

21:15

orbit and the mass of planet 9.

21:18

We don't know the phase. And so...

21:20

you know, you can draw orbits on the

21:22

nice sky all day long, all night long,

21:24

right? And you can say, well, that leaves

21:26

a lot of sky there to be to

21:29

be searched. But I'm optimistic because the LSST

21:31

is coming online this summer. and that's going

21:33

to be a game changer. It's always seemed to me,

21:35

can be surprising that, you know, it'd be so controversial.

21:37

Is it because there's so much pride associated with discovering

21:39

a planet, because there's so few of them, that is

21:42

so hotly debated, you know, they say about academics like

21:44

us, you know, the stakes are so, you know, the

21:46

stakes are so low that we have these incredibly passionate

21:48

battles. But here the stakes are high. Is that because

21:50

of the pride, is ego, is ego, is ego, is

21:52

it ego, is it ego, is it, is it, is

21:54

it ego, is it, is it, is it, is it,

21:56

is it, is it, is, is, is, is, is, is,

21:58

is, is, is, you know, you know, you, sitting at

22:00

home, whatever, drinking wine, reading the archive.

22:03

And I read, like, I saw posting

22:05

at some Rando, it was like, I'm

22:07

thinking there's a planet beyond Neptune, I,

22:10

I don't know, like, okay, moving on,

22:12

right? Like, you know, there's a natural

22:14

skepticism that you kind of gravitate to.

22:17

And, you know, I think another component

22:19

to this is that. Planets be on

22:21

Neptune have been predicted by everyone and

22:23

their brother between 1846 and now. And

22:26

it's always been wrong. Like there was

22:28

this one guy Pickering who predicted, I

22:30

don't know, like 30 of them. It

22:33

was at Harvard though, right? Yeah, I

22:35

think there's a Bayesian prior, if you

22:37

will, to this story being wrong. But

22:40

I think it's important to simply follow

22:42

the data, right? And just say, okay,

22:44

what is the data telling us, right?

22:47

meaningfully, like, well, we know that the

22:49

data is biased, right? We, like, let's

22:51

account for that, like, does it look

22:53

promising? And sometimes when, as you say,

22:56

the stakes are high, then when the

22:58

problem is important, I think it's important

23:00

to take a bit of a leap.

23:03

And even if the, you know, your

23:05

initial kind of significance is only, you

23:07

know, whatever, two sigma, right, something that

23:10

we're not something to write home about.

23:12

Like, I think it's important to pursue.

23:14

those things because the worst thing that's

23:17

going to happen is you're going to

23:19

be wrong. And like no one's going

23:21

to die, right? It's going to be

23:24

okay. So I have to I have

23:26

to always interject whenever my guest like

23:28

Constantine just did or I've had you

23:30

know five Nobel laureates sit right where

23:33

you are. Whenever they make a point

23:35

that's really crucial to the development of

23:37

good scientific habits I like to double

23:40

click on that and really enforce that

23:42

for you know half my audience has

23:44

PhDs than only has high school degrees.

23:47

And these are people that are easily

23:49

going to be influenced for good or

23:51

bad. And so when the Constantine said

23:54

just now that the stakes aren't life

23:56

or death, like you just survived a

23:58

fire at the same time. time, the

24:01

stakes for kind of what makes us

24:03

enriched as a species is the exploration.

24:05

And so what you're doing has to

24:07

be balanced, that tempered notion of not

24:10

only accepting what you want to be

24:12

true. Because it would be great, but

24:14

also to realize, yes, it's important, but

24:16

there are other things that are important

24:19

too. So I wanted to highlight that.

24:21

One question I've had as a lay

24:23

person in this field, I mean, I

24:25

love looking at planets and whatever planets

24:27

in the moon got me, planets and

24:29

whatever planets in the moon got me

24:31

into astronomy. But now what I do

24:34

is so far away from it, what's

24:36

that? Yeah, I know, I could become

24:38

your graduate. So that's something I can

24:40

return to the three body problem. seemingly

24:42

the most complicated thing in the world like

24:44

how can you predict with how many things

24:46

could be there could be a trillion objects

24:49

in the hyper belt in the or how

24:51

can you possibly predict anything I mean it's

24:53

remarkable that you even have forget about the

24:55

phase that's unknown currently but that you have

24:57

this you know five-sigma confidence bound on something

24:59

that is a localized you know probability cloud

25:02

however you want to describe it How is

25:04

that even possible when the three body promises?

25:06

You can't even do that with three bodies,

25:08

let alone trillions. Just because something is chaotic

25:10

does not mean it cannot be understood. Right?

25:12

I mean, think about weather. Okay, weather

25:14

is chaotic, right? And the level of

25:16

time is whatever, a couple days, right?

25:19

So the time for weather to forget

25:21

about its own initial conditions. And just

25:23

because that's true, doesn't mean the weather

25:25

forecast is going to be horribly wrong,

25:27

right? And so similarly. when we're

25:29

dealing with the outer solar

25:31

system, the dynamics instilled upon

25:33

the Kuiper belt, right, kind

25:35

of manifest, tells you what's

25:38

going on, not because each

25:40

particular orbit is super important,

25:43

like you should never get

25:45

obsessed over one particular KBO,

25:48

is there cumulative statistical nature

25:50

that points to what's going on.

25:52

So yeah, each one. is kind

25:55

of doing its own, you know,

25:57

stochastic thing, but cumulatively there's an

25:59

emergent. patterns. Similar with the stock

26:01

market, right? Each stock might be

26:03

quite stochastic. The cumulative behavior actually

26:05

embed some information about what's going

26:08

on. It's different as you get

26:10

a complex system and a complicated

26:12

system. I would say like building

26:14

a 787 is really complicated, but

26:16

if you do it with the

26:18

right parts the same time, every

26:20

time you get the same results,

26:22

not so with a sand pile

26:24

or with you know the weather

26:26

in San Diego or Pasadena. But

26:29

that doesn't stop people, right? So

26:31

in my field... when we come

26:33

up with an anomaly, which is

26:35

very exciting and it should herald

26:37

joy on the part of scientists,

26:39

not like depression as wrong, no,

26:41

I would say when you encounter

26:43

a flaw, it could be a

26:45

new law, right? So in the

26:47

context of what I do in

26:50

cosmology, we have something that's unknown,

26:52

like dark matter, okay? So there'll

26:54

be millions of alternative conjectures, whether

26:56

it's a different particle, it's a

26:58

field, or in fact it could

27:00

be a new modification to Newtonian

27:02

dynamics. Sure. Does that come into

27:04

play? Are there those, I'm sure

27:06

there are those, but where do

27:09

you rank these in terms of,

27:11

you know, Mond equivalents for planetary

27:13

science? And in particular, things that

27:15

aren't as abstract, like they call

27:17

them rogue planets, you know, other

27:19

trans-Neptonian objects. How do you rank

27:21

the alternative explanations? Put up the

27:23

straw man and then burn it

27:25

down for the... Okay, no, I

27:27

mean, I mean, this is an

27:30

easy exercise to do because... When

27:32

an alternative explanation comes out, right,

27:34

I try to not just, you

27:36

know, believe it, but I try

27:38

to go through and simulate it

27:40

better. Okay, an example is, you

27:42

know, there, there have been, okay,

27:44

so, uh, an example, actually, Mond,

27:46

this is not my simulation, but,

27:48

uh, David, this warning, who, as

27:51

I mentioned, was, as my collaborator,

27:53

like Mond was proposed as a,

27:55

as a, as a replacement for

27:57

planet 9 to create all of

27:59

these structures in the outer solar

28:01

system. And of course because Monde

28:03

has this tunable parameter of where

28:05

you transition from the, you know,

28:07

Newtonian to the non-Newtonian regime, right?

28:09

There were a couple of papers

28:12

that pointed towards this and this

28:14

got tested with very very high

28:16

fidelity numerical simulations and the simulations

28:18

showed that if that was the

28:20

explanation, then the orts spike of

28:22

comments, which we see very well,

28:24

would just go away. So it's

28:26

rolled out. Could it be self-gravity

28:28

of the kyper belt? This has

28:30

been an idea that kind of

28:33

floated around. We looked into this

28:35

with and dedicated a lot of,

28:37

you know, GPU time to actually

28:39

studying this and convinced ourselves. No,

28:41

and is all published like this

28:43

cannot work actually because of Neptune

28:45

scattering, etc. So each of these

28:47

alternative explanations are interesting and I've

28:49

been interested in them and I've

28:51

dedicated time to to kind of

28:54

studying them and I think it's

28:56

really important not to be religious

28:58

about your own, you know, your

29:00

own information bias is a hell

29:02

of a drug. Yeah, and so

29:04

yeah, that's what we've been doing.

29:06

So far there is no theoretical.

29:08

There's no theoretical alternative model that

29:10

I think is able to explain

29:13

the data nearly as well as

29:15

implied at 9 headphones. Hey, if

29:17

you're enjoying this, I hope that

29:19

you'll also subscribe to my Monday

29:21

Magic mailing list. We'll get to

29:23

hear some behind-the-scenes info about this

29:25

interview that I did with Constantine

29:27

and many, many more subjects that

29:29

I'd like to share with you

29:31

around the universe, the multiverse, minds

29:34

that I get exposed to. And

29:36

I love to take you for

29:38

a ride. Once you subscribe, it's

29:40

free. And if you have a

29:42

dot-e-d-u email address, you're guaranteed to

29:44

win one of these beauties, a

29:46

meteorite, a fragment of the solar

29:48

system. Before Jupiter, before planet nine,

29:50

before they were all, even glimmers

29:52

in the cosmic eye. So please

29:55

subscribe. If you don't have a

29:57

dot-e-u email address, that's OK. You

29:59

are entered to win automatically. A

30:01

competition each month. But for now,

30:03

let's get back to the episode.

30:05

professional and executive development program. Advance

30:07

your leadership skills. Craft smarter business

30:09

strategies. Build your network and transform

30:11

how you work and think to

30:13

keep your career moving forward. You'll

30:16

earn a certificate and can add

30:18

Harvard to your resume with Harvard

30:20

Professional and Executive Development. Learn more

30:22

at professional.dice.d.com. You

30:25

mentioned, you dead named, you know, the

30:27

VIR root, not, you mentioned LSST. Talk

30:29

about that. What is the excitement all

30:31

about there? What are you going to,

30:33

I've heard everything from, you know, Avilob,

30:35

who's been a many-time guest on the

30:38

show, talking about how they're going to

30:40

discover a Mumua every night. Are you

30:42

going to discover, you know, a TNO

30:44

or how is it going to revolutionize

30:46

what you do? And what Mike does.

30:48

And, you know, your collaboration, your collaboration

30:50

is a rich collaboration is a rich

30:53

one, is conceptually not that hard. You

30:55

take a picture of the night sky

30:57

and then the next day, you also

30:59

take a picture on the night sky

31:01

and you look for what has moved.

31:03

And the third day, you do that

31:06

again and you say, did that move

31:08

in a consistent manner with this being

31:10

a TNO, but I've just found one.

31:12

Okay, and for the first year or

31:14

so, all you know is how far

31:16

away it is and where it is

31:19

on the night sky. You have some

31:21

concern and inclination and inclination, but like

31:23

you don't really have a good handle

31:25

on the orbit. first. So you need

31:27

this the string of three consecutive observations

31:29

to tie together. Exactly. And you know

31:32

LSST is going to do that very

31:34

very efficiently because its entire job is

31:36

to wake up every night and kind

31:38

of look up and down the sky,

31:40

record what it saw. So it it's

31:42

going to do a lot of things

31:45

for many different fields but I think

31:47

for the outer solar system in a

31:49

way it's a really good survey. It

31:51

might not go deep enough. or might

31:53

not go north enough to find planet

31:55

9 directly, but even if it doesn't,

31:58

it will still provide... an independent check

32:00

on all of the predictions that planet

32:02

nine. Okay, so think big. The 60

32:04

minutes calls you up again. They got

32:06

a bag of cash. What's the Batijian

32:08

Observatory look like? If you could build

32:10

whatever you want, you know, money's on

32:13

an object. Where would it be? What

32:15

would it look like? Design it for

32:17

me. Oh, it's just like my laptop

32:19

in my office. And the door closed.

32:21

And the infinite supply of the impossible

32:23

coffee. Yeah. You know, I mean, I

32:26

think, you know, you know, you know,

32:28

you know, you know, you know, instruments,

32:30

right, you don't need to, you

32:32

know, really dream here, like instruments

32:35

like Subaru, like the Japanese National

32:37

Observatory and the... Yeah, they're out

32:39

there, but, you know, the thing

32:41

that has prevented... us from conducting

32:44

a surge that's really nailing down

32:46

the northern hemisphere is really the

32:48

efficiency. It's the fact that you

32:50

only get a few nights per

32:53

year, you know, and you're dominated

32:55

by the worst night that you

32:57

have of this sequence. And you

32:59

know, I wasn't an observer and

33:02

I'm still not an observer, you

33:04

know, and I'm happy about that.

33:06

But, you know, but like, I

33:08

do now, having started this, you

33:10

know, observed, like... observing about

33:12

a decade ago. I now have

33:14

this deep appreciation and gratitude for

33:17

each data point that comes up

33:19

because for especially for all the

33:21

planet nine stuff, that stuff is

33:24

up in December, like January, November

33:26

sky. And the weather in the

33:28

northern hemisphere is actually not that

33:31

good. And so that's like something

33:33

I learned. Is that it's actually

33:35

not that good. Yeah, you're like

33:38

the fogged out. There's snow, they

33:40

seeing as crap. So it's really

33:42

tough. It's really tough. It's not

33:44

as much fun as theory because

33:47

theory, as you know very well,

33:49

right? You're, you know, like you're

33:51

creating the world from scratch,

33:53

so to speak, from from axioms.

33:56

It's so much, there's so much

33:58

joy in doing that. instant gratification.

34:00

Yeah, compared with with observations, you're

34:03

just kind of at the mercy

34:05

of the telescope. the conditions, and

34:07

also what exists in the solar

34:09

system. Yeah, that's right. We're taking

34:11

a little break from the in-person

34:14

episode. I need to fold in

34:16

the actual lecture that Constantine gave.

34:18

He gave me permission to share

34:20

the lecture on Planetine gave. He

34:23

gave me permission to share the

34:25

lecture on Planet Nine, and you'll

34:27

see later Jupiter's Magnetic Field. This

34:29

is a little technical, but it's

34:32

captivated the imagination that we have

34:34

for Planet Nine. I don't want

34:36

you to miss it and he's so

34:38

grateful that Constantine gave us permission to

34:40

share this little nugget of wisdom. Then

34:43

we'll come back to the follow-up of

34:45

that in-person interview where I discuss the

34:47

fascinating aspects of how we know what

34:49

Jupiter's mass and size were some four

34:52

billion years ago. So stay tuned for

34:54

that. Now on to planet nine and

34:56

stay tuned. This is a solar system,

34:59

okay? This blue thing here is the

35:01

orbit of Neptune. Back about a decade

35:03

ago. me and my friend Mike,

35:05

inspired by work that some

35:07

of our colleagues, Chad Trujillo and

35:09

Scott Shepard did, noticed that if

35:12

you look at the most distant

35:14

orbits in the solar system, they

35:17

all swing out into sort of

35:19

the same direction and they all

35:21

are inclined with respect to the

35:24

ecliptic plane by about 20 degrees.

35:26

And we thought this was kind

35:29

of a big deal. Well, now the

35:31

data set has evolved. over the

35:33

decade. It's sort of expanded by

35:35

about a factor of three. This

35:37

is from a paper I wrote

35:40

with my friend Morby in

35:42

2017. Then you can sort

35:44

of see in 2019, there's

35:46

a little bit more objects

35:48

in 2021, more objects still,

35:50

and that's more or less

35:52

what the data set looks like

35:54

right now. So looking at this,

35:56

I think you can just like kind

35:59

of tell the more orbit swinging

36:01

out this way than another

36:03

way? And so why is

36:05

that? Well, can we invoke

36:07

that something bad happened to

36:09

the solar system when it

36:11

was forming, maybe a star

36:13

flew by, and kind of

36:15

aligned all of these objects

36:17

and we're seeing this relic?

36:19

The answer is no. Because

36:22

if you leave the solar

36:24

system alone, all of these

36:26

objects will differentially process. and

36:28

that differential procession time, the

36:30

timescale over which the structure

36:32

would become fully axi-symmetric, is

36:34

a few hundred million years.

36:36

Okay, so no. Moreover, you

36:38

see a strong correlation with

36:40

orbital stability in this plot.

36:42

Objects that are very strongly

36:44

interacting with Neptune, and in

36:47

fact, Neptune is in the

36:49

process of kicking them out

36:51

of the solar system altogether.

36:53

Here, as shown in green,

36:55

objects that are... dynamically stable,

36:57

whose pair of heli are

36:59

well enough removed from the

37:01

orbit of Neptune that nothing

37:03

happens to them are shown

37:05

in purple. And again, without

37:07

being an awesome statistician, you

37:10

can see by eye that

37:12

the purple orbits cluster together

37:14

much better than the green

37:16

ones which basically don't cluster

37:18

at all. There is a

37:20

much more... sophisticated way to

37:22

measure orbital diffusion. That's something

37:24

that's work that Gabriella Picheri,

37:26

who's a postdoc in my

37:28

group, just submitted, but maybe

37:30

I will not spend too

37:33

much time on this in

37:35

interest of time. Okay, so

37:37

if you see, you believe

37:39

what you see, and you

37:41

see these orbits, you're like,

37:43

wow, they really are clustered

37:45

together, how can that be?

37:47

Well, you need something exterior,

37:49

sensing extrinsic, extrinsic to perturb

37:51

them, to keep them confined,

37:53

and it has to be

37:55

eccentric, to break axial symmetry,

37:58

and the rest you can

38:00

compute from these types of

38:02

forward models that are just

38:04

numerical and body simulations. seeing

38:06

here is an evolutionary model

38:08

where you're starting off with,

38:10

and for scale, this is

38:12

about 30 AU, the orbit

38:14

of Neptune, you're introducing a

38:16

new planet on some highly

38:18

eccentric orbit, and you're starting

38:21

off with a rather axi-symmetric

38:23

disk of Kuiper belt objects.

38:25

And the blue orbits here

38:27

represent long period things. right,

38:29

because well they have long

38:31

period and these these golden

38:33

orbits are things that are

38:35

too short period to be

38:37

strongly affected by planet nine

38:39

induced dynamics. So it takes

38:41

a couple billion years for

38:44

a pattern to emerge, but

38:46

right about now we're starting

38:48

to see how the anti

38:50

aligned direction with respect to

38:52

the orbit of the introduced

38:54

perturbar is kind of starting

38:56

to get preferred, right? There

38:58

are more objects hanging out

39:00

here. You guys also see

39:02

this, right? Like I'm not

39:04

alone. Okay, that's good. Problem

39:06

if I was the only

39:09

one. Okay, why is this

39:11

happening? Right? Why anti-line? Well,

39:13

as you can see, occasionally,

39:15

objects will process through orbital

39:17

alignment. and when they process

39:19

through orbital alignment, their eccentricity

39:21

reaches a peak and their

39:23

orbits get jammed into the

39:25

orbit of Neptune, which then

39:27

scatters them out of the

39:29

solar system. Okay, so this

39:32

is kind of a survival

39:34

technique, if you will, of

39:36

long period, Kuyper belt objects.

39:38

Okay. And the same thing

39:40

largely. remains true also for

39:42

the plane. Okay, so if

39:44

we go into 3D, we'll

39:46

find that the surviving objects

39:48

also get tilted away from

39:50

the plane, the ecliptic plane,

39:52

by effectively bending of the

39:55

Laplace plane, by gravity of

39:57

planet 9. Okay, there's also

39:59

very high inclination. dynamics that

40:01

gets excited. Okay, good. So,

40:03

if you believe that this is the

40:05

case, right, then you can compute what

40:07

the best kind of fit

40:10

planet nine parameters are, and

40:12

they turn out to be about

40:14

five Earth masses with an orbital

40:16

period of about 10,000 to 20,000

40:19

years. This is a thing 500-a-U

40:21

in terms of seven major

40:23

axis, and an eccentricity of about

40:26

point three. inclination of about

40:28

20 degrees. That's kind of what

40:30

you what you get from.

40:32

Now there's been some discussion in

40:34

the literature about whether or

40:37

not this is actually real, right?

40:39

People talk about, well, you

40:41

know, what if all of

40:43

this is a conspiracy of

40:45

observational biases that together make

40:47

this pattern? And we've, you

40:49

know, participated in that debate. I

40:52

would argue that the false alarm

40:54

probability here is 0. 0.2 percent.

40:56

But that's not what I want

40:59

to talk about, okay, because

41:01

I want to leave that question

41:03

for Vera Ruben. Instead, what

41:05

I want to think about

41:07

is a distinct, you know, a distinct

41:10

process, namely, up until

41:12

now, right, I've been asking

41:14

you to focus on these

41:16

objects that are very, very

41:18

stable. These things that are

41:21

removed from the orbit of

41:23

Neptune, so they have, they're

41:25

corralled by planet nine's gravity,

41:27

they hold the footprint or

41:29

thumbprint of secular interactions with

41:31

planet nine. And also if you

41:33

were paying attention to the previous

41:35

slide, you saw how we started

41:37

out with lots and lots of

41:39

objects, and then many of them

41:41

disappeared, right, because they got jammed

41:43

into the orbit of Neptune. So what

41:46

these calculations tell you is

41:48

that if Planet 9 really

41:50

exists, it should also, in

41:52

addition to doing this

41:54

confining business, drive a

41:56

steady flux of long

41:58

period object. that cross the

42:01

orbit of Neptune. And well,

42:03

here are some numerical simulations

42:05

of the chaotic evolution of

42:08

the perhealian distances, for example,

42:10

where you see that happening,

42:12

where perhealia dip below Neptune,

42:14

right? And then at the

42:17

end of, you know, as

42:19

the solar system, at the

42:21

moment when we're observing now,

42:23

they're being just like jammed

42:26

into the space in between

42:28

the giant planets. Well,

47:08

with the name of your price

47:10

tool from Progressive, you can find

47:12

options that fit your budget and

47:14

potentially lower your bills. Try it

47:16

at progressive.com. Progressive Casualty Insurance Company

47:19

and affiliates. Price and coverage match

47:21

limited by state law. Not available.

50:00

kind of theoretically, you know,

50:02

self-consistent, but nevertheless vague picture

50:04

of first you form a

50:06

core, then this core slowly

50:08

accretes an atmosphere that's hydrostatic,

50:10

and once the mass of the atmosphere

50:12

becomes as big as the core itself,

50:15

then you enter a runaway phase of

50:17

accretion where the planet grows very fast

50:19

to Jupiter Mouse. Okay, so when did

50:21

that... Take place exactly like how what

50:24

was the state of Jupiter like

50:26

you know at some point after

50:28

the Sun's formation other than now

50:30

Yeah, we don't know and so

50:32

what this new work? Demonstrates is

50:34

that there's actually a record

50:36

of how Jupiter evolved about

50:38

four million years after formation

50:40

of the first solids and the

50:42

solar system that's embedded within

50:44

the orbits of the tiny

50:46

satellites that live inside of

50:48

Iowa. Okay, so there's, there's, sorry, satellites

50:50

inside of Iowa, yeah, inside the

50:53

orbit of Iowa, there are, there

50:55

are, yeah, everybody, everybody always forgets

50:57

that these exist, but actually the

50:59

first one in Maltia was discovered

51:01

by Barnard, yeah, and like, 1826,

51:04

maybe 90, but like, he must

51:06

have had crazy good vision, right,

51:08

because this satellite is like 80

51:10

kilometers across. and it orbits at

51:12

only a couple two and a

51:15

half or so jovian radio and

51:17

radii and there's another one called

51:19

Thebe that's slightly further out and as

51:21

it turns out the orbital inclinations of

51:23

these moons store a record of where

51:26

IO started out how it migrated out

51:28

tidily and from this you can infer

51:30

a lot. I have a little shameful

51:32

detail, a secret, a metric to reveal

51:34

to you, which is that only about

51:36

a third of you that are watching

51:39

and enjoying or listening to this podcast

51:41

or watching it on YouTube are actually

51:43

subscribed and following me on those platforms.

51:45

And it's quite a shame because we

51:47

have so many cool episodes coming up

51:49

with the actual man who killed Pluto, Mike

51:52

Brown, it's coming up. You don't want to

51:54

miss it. So please do subscribe or follow

51:56

it wherever you're watching or listening to it.

51:58

I guarantee it's worth your time. If you

52:00

wouldn't mind doing me a favor,

52:02

an astronomical favor, you can't have

52:05

your own constellation. Those are set.

52:07

There's only 88 of those. But

52:09

you could make your own asterism,

52:11

a collection of five stars, hopefully,

52:13

where you can review the podcast

52:15

if you're listening on audio. So

52:18

please do that on Apple or

52:20

Spotify. It really means a lot

52:22

to me, and it really does

52:24

help us boost the... ratings visibility

52:26

and quality and caliber of the

52:29

productions have really upped it better

52:31

cameras better sound better lighting and

52:33

I know that you'll appreciate it

52:35

so please do do that and

52:37

I hope you will see it

52:40

will pay off course it's free

52:42

so doesn't really cost you anything

52:44

please do that now back to

52:46

the episode the basic idea is

52:48

that from Iowa's orbital record, right?

52:51

You can also use, you know,

52:53

conservation laws, conservation of angular momentum

52:55

of the spin of Jupiter, etc.

52:57

to read off, what was it

52:59

like when the gas just evaporated?

53:02

And the answer is, it was

53:04

twice as big as it is

53:06

now, okay? And it glowed at

53:08

about 1,200 Kelvin. So it's a

53:10

almost brown door for even harder

53:13

than a brown door. Oh, yeah,

53:15

initially, yeah, absolutely. And so it

53:17

was fusion. I mean, could I

53:19

have fusion at that scale? No,

53:21

no. Because it didn't have the

53:24

interior temperature, even like D.T. Fusion

53:26

requires 70,000 Kelvin. It's like at

53:28

50 at the so close, but

53:30

no cigar. But once you know

53:32

the interior state, then you can

53:35

infer the magnetic field. Why? Because

53:37

as it turns out. rapidly spinning

53:39

spherical fully convective astrophysical dynamos all

53:41

fall in this regime of having

53:43

a equal partition-like behavior where the

53:46

kinetic energy of the convection is

53:48

so like rovy squared of convective

53:50

you know motion within the planet

53:52

is balanced by the magnetic energy

53:54

density, B squared over 2m, naught.

53:57

And that comes from the fact

53:59

that convection is the thing that's

54:01

generating the field. And there's kind

54:03

of, there's a bucket to put

54:05

the energy in. And so from

54:08

like the solid, rocky planets, right?

54:10

Right, right. Well, and so that

54:12

gives you a couple hundred gals

54:14

as the field of the primordial

54:16

Jupiter. And once you know the

54:19

radius and the field, and you

54:21

know where IO was parked, you

54:23

can actually also infer from that

54:25

the accretion rate. of gas that

54:27

Jupiter was experiencing right before the

54:30

gas went away. And that turns

54:32

out to be a Jupiter mass

54:34

per million years. So all of

54:36

these things are actually not surprising

54:38

numbers, but it is a model

54:41

independent way to infer them. And

54:43

it's like stored in the orbits

54:45

of these timing cells. And are

54:47

there implications for the survival, I'm

54:49

just thinking right now about Earth.

54:52

And as you said, Jupiter is

54:54

the architect, but it's also like

54:56

a bodyguard. We saw a shoemaker

54:58

levee, you probably weren't. baby face

55:00

petitian they call you you know

55:03

there's this notion of as a

55:05

bodyguard absorbing stuff so if it

55:07

was eight times bigger in a

55:09

volume it was eight times more

55:11

massive roughly it's just you know

55:14

whatever i'm a cop experimentalist right

55:16

does that mean it was even

55:18

more efficient soaking up the meteorites

55:20

the meteors that would have impacted

55:22

earth comments transductionian objects planet objects

55:25

planet 27 could it have been

55:27

more you know of a bodyguard

55:29

than it already was and allowed

55:31

life I'm trying to get to

55:33

yeah yeah I think certainly certainly

55:36

compared to the work it does

55:38

now, like right now the load

55:40

is low, because there's not that

55:42

much stuff coming in from the

55:44

outer solar system. So just a

55:47

flux of transymptuenian objects, you know,

55:49

becoming centaurs and being kicked around

55:51

by the giant planets, and then

55:53

eventually reaching Jupiter to become Jupiter

55:55

Famicombez, that flux is nothing compared

55:58

to what it was immediately after

56:00

the gas went away in the

56:02

primordial solar system. When that happened,

56:04

the solar system was encircled by

56:06

20 Earth masses of planetesimal. And

56:09

that stuff all had to get

56:11

scattered out. So Jupiter was kind

56:13

of working overtime in the first.

56:15

clipper, Jews, all these other emissions,

56:17

will they tell you anything? Are

56:20

they more the outer moons so

56:22

they don't tell you as much

56:24

about IEO? They will tell us

56:26

a lot about the composition, the

56:28

geophysics and the geochemistry of the

56:31

satellites. You know, I work a

56:33

little bit on satellite formation, a

56:35

couple papers, and I'm really excited

56:37

because that's going to kind of

56:39

take the real... constraints of just

56:42

being there and kind of taking

56:44

a look at what this looks

56:46

like to the next level. For

56:48

this problem I don't think it

56:50

will matter that much, but you

56:53

know, I always don't think things

56:55

matter that much until they do.

56:57

So, you know. And then Leslie,

56:59

whatever his name is, shows up

57:01

on your doorstep. Her name is,

57:04

talk about the calculations that surprised

57:06

me in this paper, the consequences

57:08

of thermodynamics. I didn't initially see

57:10

that there'd be any connection between

57:12

the entropy, what you call the

57:15

cold-store and a hot-store. What are

57:17

they first explained? How is entropy

57:19

relevant to, you know, these calculations,

57:21

you know, So can I tell

57:23

you a quick story that's like

57:26

one of my favorite moments from

57:28

undergrad is I had a I

57:30

had a professor in thermo who

57:32

said on lecture one is like

57:34

you have been all misled about

57:37

entropy you've been told that entropy

57:39

is a measure of disorder in

57:41

the universe or in your system

57:43

and that's just like that's just.

57:45

Like, forget that. Okay, in this

57:48

class we're going to really learn

57:50

what it is. But before you

57:52

do, you have to first learn

57:54

quantum statistics, then from it, you're

57:56

going to get classical statistics, so

57:59

there's like kind of three weeks

58:01

of prep that you have to

58:03

do before you really understand what

58:05

entropy is. And then came the

58:07

day when he was like, today

58:10

you will learn what entropy really

58:12

is, and it'll finally make full

58:14

sense to you. Everyone's, you know,

58:16

super excited. of the partition function.

58:18

Do you understand? So that's my

58:21

answer. That's how it's all connected.

58:23

But to maybe bring it back

58:25

to. something a little bit more

58:27

practical, right? Jupiter, by virtue of

58:29

being a convective planet, is also

58:32

very nearly isentropic. Even though temperature

58:34

of course goes up as you

58:36

go into the deep interior, if

58:38

you were to grab a patch

58:40

of gas from, I don't know,

58:43

halfway into Jupiter's radius and slowly

58:45

move it back up to the

58:47

surface, it would have the same

58:49

temperature as the surface, right, or

58:51

that's the convective boundary, effectively speaking,

58:54

right? And so the entropy... is

58:56

a much better number than temperature

58:58

because it's the thing that defines

59:00

the entire kind of curve of

59:02

the interior profile. It's a record

59:05

of the trace of the past.

59:07

Exactly. And so quoting a number

59:09

like whatever, 10.5, right, KB per

59:11

barion, right, that. tells you how

59:13

hot Jupiter is, not just at

59:16

the surface, but how hot it

59:18

is in the interior. It kind

59:20

of gives you the full picture.

59:22

Now, this hot start versus cold

59:24

start problem is ultimately comes down

59:27

to the problem of the shock.

59:29

So when you're forming the planet,

59:31

okay, and the planet is accreating,

59:33

and gas is falling on it,

59:35

right? If you imagine taking a

59:38

balloon of gas and smacking it

59:40

against another balloon gas, you do

59:42

it slowly, it just absorbs, right?

59:44

If you do it much faster

59:46

than the speed of sound, it

59:49

bounces. And so you get some

59:51

energy release. So there's this question,

59:53

which is not a, this is

59:55

theoretically difficult question to answer, of

59:57

how much energy is actually injected

1:00:00

into Jupiter when it decretes mass,

1:00:02

and how much of it just

1:00:04

bounces off and radiates away? this

1:00:06

new paper suggests is that actually

1:00:08

most of it gets injected into

1:00:11

Jupiter. As you accrete, a large

1:00:13

fraction of the luminosity that the

1:00:15

planet, you know, exudes, is coming

1:00:17

from material that's being injected. That

1:00:19

was awesome. That was a kind

1:00:22

of grand overview of what we

1:00:24

know about Jupiter and how we

1:00:26

know it. Now here's a deep

1:00:28

dive from the slides from the

1:00:30

presentation that we heard Constantine just

1:00:33

give to us here in person

1:00:35

at UCSD. This is again a

1:00:37

real and rare treat for you

1:00:39

and me to experience the top

1:00:41

performer in this field. It's like

1:00:44

having Steph Curry teach you out

1:00:46

to do jump shots. So it's

1:00:48

a master class from an expert,

1:00:50

perhaps again, the foremost expert in

1:00:52

the world on the... properties of

1:00:55

our early solar system, including the

1:00:57

second most important planet, okay, fine,

1:00:59

Jupiter. So now we're going to

1:01:01

go into that slide and then

1:01:03

we'll come back to the very

1:01:06

end of the interview and then

1:01:08

we'll have homework and takeaways for

1:01:10

you. So we've got... all these

1:01:12

circles, right, these are all planets

1:01:14

color coded by the type of

1:01:17

star that they orbit. And what

1:01:19

I've done here is on the

1:01:21

y-axis, I've done a slight variation.

1:01:23

Usually people just to show the

1:01:25

mass, I've normalized the mass by

1:01:28

the mass of the central objects

1:01:30

so that I could also overplot

1:01:32

the population of solar system satellites

1:01:34

as these rectangles. So naively, just

1:01:36

like without knowing anything. you can

1:01:39

kind of tell that the population

1:01:41

of giant planet satellites fits nicely

1:01:43

with this cloud of subjovian extrasolar

1:01:45

planets that as it turns out

1:01:47

is the dominant outcome of planet

1:01:50

formation in the galaxy. Now there

1:01:52

are many patterns that are being

1:01:54

studied about this population day and

1:01:56

I would say one of the

1:01:58

most striking things is that They're

1:02:01

all kind of right here. This

1:02:03

is a log scale, so it's

1:02:05

easy to say right here and

1:02:07

kind of cover a lot of

1:02:09

space. But if you focus on

1:02:12

this histogram here, it shows you

1:02:14

a histogram of the shortest orbital

1:02:16

period of a planet in a

1:02:18

given system. And there is clearly

1:02:20

some peak, right, that lives between

1:02:23

a period of one day and

1:02:25

10 days. How do we understand

1:02:27

this being like, why should this

1:02:29

be? Where does this come from?

1:02:31

Well, in general, we think these

1:02:34

planets, when they form, interact with

1:02:36

the protoplanetary disk, where they form,

1:02:38

right? And they do so principally

1:02:40

by raising wakes. within the gas

1:02:42

and these wakes, sorry, gravitationally pulled

1:02:45

back on the planet. And so

1:02:47

if you put a planet somewhere

1:02:49

within the disk through this interaction,

1:02:51

which is creatively called type one

1:02:53

migration, there's also type two. So

1:02:56

type one migration, the orbit just

1:02:58

decays and it decays all the

1:03:00

way down to the place where

1:03:02

the disk ends. And the disk,

1:03:04

right, the protoplanetary disk, has a

1:03:07

cavity because magnetosphere of stars tend

1:03:09

to carve out this cavity. And

1:03:11

this is sort of a well-known

1:03:13

and well-appreciated feature of protoplanetary disks.

1:03:15

Okay. Could it be a selection

1:03:18

effect? Let's go back. It's easier

1:03:20

to find stuff this way. Okay.

1:03:22

So. No,

1:03:24

no, it's easier to find

1:03:27

stuff this way. So the

1:03:29

fact that there's a drop-off

1:03:31

here, here is real. Out

1:03:33

here, is there selection in

1:03:35

fact? Absolutely. And people do

1:03:38

very careful modeling of asking

1:03:40

the question of like, is

1:03:42

this fall-off real? And the

1:03:44

short answer is real. There's

1:03:47

really a turnover in the

1:03:49

knee of the occurrence distribution.

1:03:51

Okay. So there's, if you

1:03:53

kind of accept that. Protoplanetary

1:03:56

disks are truncated by magnetospheres

1:03:58

like people generally accept that

1:04:00

to be true and if

1:04:02

you accept that this interaction

1:04:05

leads you to decay to

1:04:07

the inner disk, then you're

1:04:09

presented with a bit of

1:04:11

a puzzle. And this puzzle

1:04:13

I can highlight by going

1:04:16

to rent three systems, which

1:04:18

orbit three different types of

1:04:20

stars. Kepler 256 has some

1:04:22

planets, they orbit a star

1:04:25

of a solar mass, and

1:04:27

the innermost period is one

1:04:29

and a half days. And

1:04:31

then if you go an

1:04:34

order of magnitude down, the

1:04:36

Trappist 1 system, which is

1:04:38

a very famous exoplanet system,

1:04:40

in part because it's called

1:04:43

Trappist, and you ask, what

1:04:45

is the innermost orbital period?

1:04:47

It's also a half, well,

1:04:49

one and a half days.

1:04:52

If you go another two

1:04:54

orders of magnitude down and

1:04:56

ask, where's Jupiter? Like, where

1:04:58

is IO orbit? It's sort

1:05:00

of also one and a

1:05:03

half days. So I don't

1:05:05

want to make the impression

1:05:07

that one and a one

1:05:09

and a one and a

1:05:12

half days is... you know,

1:05:14

absolutely the critical number, but

1:05:16

the order of magnitude is

1:05:18

kind of conserved, even though

1:05:21

the mass of the central

1:05:23

body changes by orders of

1:05:25

magnitude. So how can this

1:05:27

be, right? How can there

1:05:30

be within this context a

1:05:32

deep level, deep like state

1:05:34

level of conspiracy where all

1:05:36

disks get truncated at an

1:05:38

orbital period of only a

1:05:41

few days? So let's think

1:05:43

about how this can be?

1:05:45

Well, first of all, the

1:05:47

physics of truncation of protoplanetary

1:05:50

disks has been understood since

1:05:52

at least 1979, like literature

1:05:54

in Neutron Stars, my Gaussian

1:05:56

lamb, was really the first

1:05:59

to point out that you

1:06:01

can compute this radius of

1:06:03

the magnetosphere cavity by equating

1:06:05

the magnetic pressure scale. to

1:06:08

the accretionary ram pressure scale.

1:06:10

So you can do that.

1:06:12

assuming a dipole field, you

1:06:14

know, magnetic pressure as usual

1:06:17

is B squared over two

1:06:19

new naught, and for kind

1:06:21

of spherical free fall, RAM

1:06:23

pressure, row B squared, can

1:06:25

be re-expressed in terms of

1:06:28

the disk M. Dot. Okay,

1:06:30

so these two things, the

1:06:32

radius at which these two

1:06:34

things equal is where you

1:06:37

cut the disk. Okay, so

1:06:39

somehow, right, the radius changes

1:06:41

with. the essential mass, right?

1:06:43

But the frequency remains the

1:06:46

same. So how can we

1:06:48

have this? Well, let's let's

1:06:50

compute, let's construct a very

1:06:52

simple model. So the simplest

1:06:55

thing, the simplest scaling that

1:06:57

you can imagine for the

1:06:59

accretion rate of prooplanetary or

1:07:01

just like disk astrophysical disks

1:07:03

is that the rate of

1:07:06

accretion will scale with the

1:07:08

mass of the central body.

1:07:10

is actually quite fuzzy. This

1:07:12

might be to the one

1:07:15

power, this might be to

1:07:17

the two power. Both are

1:07:19

consistent with the available data,

1:07:21

but for the simplicity, let's

1:07:24

choose this linear relationship. Okay,

1:07:26

so then we'll replace the

1:07:28

M dot here with something

1:07:30

that goes as M. Okay,

1:07:33

what about the magnetic field?

1:07:35

Well, for rapidly rotating, fully

1:07:37

convective, astrophysical dynamos, there exists

1:07:39

an important scaling law that

1:07:41

tells you that magnetic energy

1:07:44

density be squared over two

1:07:46

mu naught, goes roughly as

1:07:48

the kinetic energy density of

1:07:50

convection. And then through mixing

1:07:53

length theory, you can relate

1:07:55

this in the usual way

1:07:57

to the heat flux. This

1:07:59

is a reason, by the

1:08:02

way, I'm telling you all

1:08:04

this. I'm not just randomly

1:08:06

making stuff up. This is

1:08:08

all going to connect back

1:08:11

to Jupiter momentarily, but I'm

1:08:13

having... fun first with extra

1:08:15

solar planets. Okay, so this

1:08:17

scaling law between the field

1:08:20

strength and the luminosity of

1:08:22

stars is a pretty well

1:08:24

established thing and it connects,

1:08:26

you can actually connect the

1:08:28

geodinemo, the jovian dynamo, M.

1:08:31

dwarfs, all on the same

1:08:33

curve. Now what about the

1:08:35

radius? Well, remember, early on,

1:08:37

while things are encircled by

1:08:40

protoplanet by disks. Stars are

1:08:42

contracting, roughly as just Kelvin

1:08:44

Helmholds contraction, and this is

1:08:46

a well-known result, that the

1:08:49

radius then is also just

1:08:51

expressed in terms of the

1:08:53

heat flux. And as it

1:08:55

turns out, if you put

1:08:58

all of these things together,

1:09:00

right, you can derive an

1:09:02

equation for the frequency, orbital

1:09:04

frequency, at which the disk

1:09:06

will be truncated, and all

1:09:09

of the dependence on the

1:09:11

mass. goes away. And all

1:09:13

of these various constants that

1:09:15

appear in the scaling laws

1:09:18

are there, but they come

1:09:20

in at a sublinear power.

1:09:22

So you kind of get

1:09:24

this two pie over three

1:09:27

days orbital frequency as a

1:09:29

relatively universal outcome of disc

1:09:31

truncation. And I would argue

1:09:33

that the fact that IO

1:09:36

and Trappist 1 and all

1:09:38

the usual Kepler XO planets

1:09:40

all orbit. In a matter

1:09:42

of a few days, it's

1:09:45

just a reflection of the

1:09:47

interplay of these mechanisms, right?

1:09:49

Convective dynamo generation, just regular

1:09:51

disc accretion, just regular disc

1:09:53

accretion, and Kelvin helipons contraction.

1:09:56

So as we enter the

1:09:58

age of characterization of circumplanetary

1:10:00

disks, for which PDS 70C

1:10:02

is the poster child, here's

1:10:05

PDS 70C. There's clearly a

1:10:07

circumn planetary disk here. If

1:10:09

you don't see it... Look

1:10:11

again. Okay. It's there. Okay.

1:10:14

This blob is a disk.

1:10:16

Okay. So I would argue

1:10:18

that as we enter discover

1:10:20

more of these things in

1:10:23

the age of Alma You

1:10:25

know we will find that

1:10:27

these two will be truncated

1:10:29

at a period of on

1:10:31

the order of a few

1:10:34

days and that's the preamble

1:10:36

and the reason I wanted

1:10:38

to tell you this is

1:10:40

because much of the same

1:10:43

physics that I just mentioned

1:10:45

will come back momentarily when

1:10:47

we talk about Jupiter. Okay,

1:10:49

so why do we care

1:10:52

about Jupiter? First of all,

1:10:54

every person interested in celestial

1:10:56

mechanics to have ever lived

1:10:58

has concluded that the solar

1:11:01

system is composed of the

1:11:03

Sun, Jupiter. and other things.

1:11:05

In fact, if you read

1:11:07

like the textbook of Arnold,

1:11:09

not Arnold, Schwarzenegger, but a

1:11:12

different Arnold, like the mathematician,

1:11:14

he kind of says this

1:11:16

and he says, okay, so

1:11:18

everything else we'll do in

1:11:21

this textbook is basically going

1:11:23

to be in this framework

1:11:25

of the restricted circular three-body

1:11:27

problem. Also, you know, as...

1:11:30

our understanding of how the

1:11:32

solar system came into existence

1:11:34

has sharpened up, it's become

1:11:36

clear that actually the formation

1:11:39

of Jupiter played a great,

1:11:41

you know, defining role in

1:11:43

setting the large-scale architecture of

1:11:45

our solar system. Perhaps even

1:11:48

the fact that the terrestrial

1:11:50

planets are so low-mass is

1:11:52

connected to the fact that

1:11:54

Jupiter formed. And these days,

1:11:56

and by the way, Jupiter-like

1:11:59

planets are not a given.

1:12:01

Right, Jupiter-like planets occur at

1:12:03

around 10 to 15% of

1:12:05

sun-like stars, much less common

1:12:08

for lower metallicity, lower mass

1:12:10

stars. So by virtue of

1:12:12

having giant planets in the

1:12:14

first place, our solar system

1:12:17

kind of... already scores at

1:12:19

least a B plus. Okay,

1:12:21

so it's pretty good planetary

1:12:23

system. Yes. Oh,

1:12:31

yeah, good question. So is

1:12:33

it simply an observational bias?

1:12:35

I would say at this

1:12:37

point, no, because some of

1:12:39

the, I mean, that number

1:12:41

at this point comes from

1:12:43

the California Legacy Survey, which

1:12:45

has been going on for

1:12:47

nearly 40 years, right? So,

1:12:49

of course, with ever increasing

1:12:51

precision, but yeah, at this

1:12:53

point in terms of. period,

1:12:55

we kind of go beyond

1:12:57

Saturn. And there appears to

1:12:59

be a drop-off in their

1:13:01

occurrence rate before the bias

1:13:03

really sets in. So there

1:13:05

seems to be, like, a

1:13:07

couple A-U is the peak

1:13:09

of where giant planets occur,

1:13:11

and they're much more rare

1:13:13

interior and exterior to that.

1:13:15

So it's kind of this

1:13:17

log galsion distribution. Okay. So.

1:13:20

Today, we know quite a

1:13:22

bit about Jupiter itself. We've

1:13:24

got the Juno mission, which

1:13:26

orbits Jupiter. One of the

1:13:28

goals of the Juno mission

1:13:30

was to measure the gravitational

1:13:32

harmonics out to degree like

1:13:34

12,000. It's not really 12,000,

1:13:36

but it's some very, very

1:13:38

high degree. I think they

1:13:40

have like a Lejeune Polynom

1:13:42

out to J14, right? So

1:13:44

just crazy, crazy, good understanding

1:13:46

of the Jovian. you know,

1:13:48

the jovian gravitational field, there's

1:13:50

all this understanding of what's

1:13:52

in the jovian atmosphere, and

1:13:54

I would argue that by

1:13:56

comparison, our understanding of how

1:13:58

Jupiter formed, can be summarized

1:14:00

in this plot from 1996,

1:14:02

and this is still more

1:14:04

or less the state of

1:14:06

the art. So let's go

1:14:08

through this plot. What is

1:14:10

it showing us? Well, first

1:14:12

of all, on the x-axis,

1:14:14

it's showing us time in

1:14:16

millions of years. And on

1:14:18

the y-axis, it's showing us

1:14:20

mass. So this plot can

1:14:22

be separated out into three

1:14:24

distinct phases, which are named

1:14:26

phase one, phase two, and

1:14:28

phase three. Okay, phase one,

1:14:30

which is this phase, corresponds

1:14:32

to the formation of the

1:14:34

core of Jupiter. It's like

1:14:36

from Jupiter gravity data, we

1:14:38

know that there's about 25

1:14:40

Earth masses of heavy elements

1:14:42

inside Jupiter. They're not concentrated

1:14:44

in a straight up solid

1:14:46

core. They're kind of distributed

1:14:48

in a fuzzy core, but

1:14:50

we know that the core

1:14:52

is relatively deep-seated. So once

1:14:54

this core forms, then we

1:14:56

have a protracted period of

1:14:58

steady gas accretion where this

1:15:00

core acquires a hydrostatic envelope.

1:15:02

that slowly grows in mass.

1:15:05

Okay, it grows simply by

1:15:07

cooling down. In fact, Eve

1:15:09

has a paper about this,

1:15:11

right? It's just like, cools

1:15:13

down, so when it cools

1:15:15

down, it contracts a little

1:15:17

bit, letting in more gas

1:15:19

into the hill sphere. That's

1:15:21

the basic mechanism. And once

1:15:23

the gas accretion allows the

1:15:25

atmosphere to become as massive

1:15:27

as the core itself, this

1:15:29

process accelerates into a phase

1:15:31

of runaway accretion during which

1:15:33

you grow up and graduate

1:15:35

from sort of 20, 30

1:15:37

earth masses all the way

1:15:39

up to the 300 earth

1:15:41

masses that is Jupiter in

1:15:43

a short amount of time.

1:15:45

In fact, in these 1D

1:15:47

models, the accretion rate goes

1:15:49

as something like mass to

1:15:51

the four-thirds power, and so

1:15:53

you reach infinite mass and

1:15:55

finite time. And the way

1:15:57

that you explain that Jupiter

1:15:59

is not infinitely massive is

1:16:01

at some point you just

1:16:03

have to turn off the

1:16:05

code, right? Like when it

1:16:07

goes through Jupiter mass, you

1:16:09

shut that sucker down, okay?

1:16:11

Shut off the gas. So

1:16:13

this is from 1996. right

1:16:15

nickel back hadn't even made

1:16:17

it big okay like that's

1:16:19

how old this plot is

1:16:21

right this is from 2019

1:16:23

and modern kind of 3D

1:16:25

calculations more or less look

1:16:27

like this and they have

1:16:29

had a illuminating effect in

1:16:31

quantifying how the hydrodynamics of

1:16:33

gas occurs when you have

1:16:35

a young planet that is

1:16:37

embedded within a protoplanetary disk.

1:16:39

And it's very, very interesting.

1:16:41

But when it comes to

1:16:43

answering the question of like,

1:16:45

what happens to the, what's

1:16:47

going on at the planetary

1:16:49

scale, these models basically have

1:16:52

no resolving power in part

1:16:54

because their softening length is

1:16:56

about this big, okay, 0.1

1:16:58

heels fears. And it's frustrating

1:17:00

because like I would like

1:17:02

to know. how Jupiter formed.

1:17:04

Now, I'm not in the

1:17:06

astronomy department at Caltech, I'm

1:17:08

in planetary science, which is

1:17:10

part of geological and planetary

1:17:12

sciences. A lot of my

1:17:14

colleagues are geologists. And one

1:17:16

of the things that you

1:17:18

learn about a geologist is

1:17:20

like, if you go out

1:17:22

into a field with a

1:17:24

geologist, a geologist will kind

1:17:26

of walk around for a

1:17:28

while, pick up a rock,

1:17:30

kind of look at it,

1:17:32

put it back down, pick

1:17:34

up another one to kind

1:17:36

of smell it, and be

1:17:38

like. that mountain then should

1:17:40

definitely feel like formed 50

1:17:42

million years ago. Like I

1:17:44

just know it. How did

1:17:46

you know? It's like you

1:17:48

just know. Okay. So I

1:17:50

always kind of feel jealous

1:17:52

that I can't just like

1:17:54

look at a rock and

1:17:56

just know how. Jupiter formed,

1:17:58

except for I think you,

1:18:00

like there is a chance,

1:18:02

okay, there is a chance.

1:18:04

So Jupiter, and this is,

1:18:06

by the way, a beautiful

1:18:08

JWST image of Jupiter, so

1:18:10

IO is, it's not in

1:18:12

the image, it's further out.

1:18:14

If you look at Jupiter

1:18:16

close in, it's orbited by

1:18:18

a series of rocks, and

1:18:20

these are rocks that are

1:18:22

maybe 80 kilometers across, and

1:18:25

people always forget that these

1:18:27

rocks exist. Okay, in fact, this

1:18:29

one, Amaltia, was discovered by Barnard,

1:18:31

of the Barnard star fame in

1:18:33

his paper. He speculated about the

1:18:36

kinds of aliens that live on

1:18:38

Amaltia, which is pretty fun to

1:18:40

read. But there's now, as it

1:18:42

turns out, there's four of them.

1:18:44

There's Amaltia here, there's Stevie, which

1:18:46

is off the image, but you

1:18:49

can see some of the light

1:18:51

there, and there's a couple other

1:18:53

really, really, really, really tiny rocks

1:18:55

that actually create the Jovian rings.

1:18:58

And these rocks, even though,

1:19:00

by the looks of it, they

1:19:02

orbit exactly in

1:19:04

the plane, in the

1:19:07

kind of equatorial plane

1:19:09

of Jupiter, that correspondence

1:19:12

is in fact not

1:19:14

precisely exact. Okay? Amaltia

1:19:16

is inclined with respect

1:19:19

to the Jovian plane

1:19:21

by 0.39 degrees and

1:19:23

Phoebe. is inclined with

1:19:25

respect to the Jovian

1:19:28

equatorial plane by 1.1

1:19:30

degrees. People in astronomy

1:19:33

would like to say what's 1.1

1:19:35

degrees among friends, that's like

1:19:37

zero, but I would argue

1:19:40

that these are in fact

1:19:42

very very meaningful numbers. Why

1:19:44

are they meaningful numbers? They're

1:19:47

meaningful numbers because they in

1:19:49

fact hold the record of

1:19:52

IO's tidal regression. Yeah

1:20:00

What do you see? Iow's

1:20:02

foot brown? Oh, this? Oh,

1:20:04

okay. So Iow is heavily

1:20:06

volcanic, right? And so it's

1:20:08

always like the plasma Taurus

1:20:11

is part of the plasma

1:20:13

Taurus is accreating onto the

1:20:15

following the Jovian field lines.

1:20:17

Okay. So given what I've

1:20:20

told you in the first

1:20:22

few slides. about extra solar

1:20:24

planets. And the fact that

1:20:26

almost certainly qualitatively the same

1:20:29

thing unfolded in the Jovian

1:20:31

system, namely the satellites formed

1:20:33

somewhere by type one torques,

1:20:35

they migrated and parked near

1:20:37

the inner edge of the

1:20:40

circum Jovian disk. And that's

1:20:42

actually why they're in a

1:20:44

four to two to one

1:20:46

residence. Okay. So at the

1:20:49

time when the disk is

1:20:51

right about... ready to dissipate.

1:20:53

The picture is as follows.

1:20:55

You have i.o, Europa, Ganimied,

1:20:58

Calisto somewhere here, and the

1:21:00

two rocks, Amaltia and Thebe,

1:21:02

are inside the magnetospheric cavity.

1:21:04

Why are they inside the

1:21:07

magnetospheric cavity? It's because, well,

1:21:09

they're too massless to experience

1:21:11

meaningful. type one torques, they

1:21:13

are just shepherded inwards by

1:21:15

residences with with Iow. I

1:21:18

can dwell on that a

1:21:20

little bit longer, but for

1:21:22

now, just trust me, they

1:21:24

were inside the inner edge.

1:21:27

Now, then the disk photo

1:21:29

evaporates at some point. Typical

1:21:31

disks live for about three

1:21:33

million years. I argue that

1:21:36

the solar system's disk lived

1:21:38

a little bit longer. We'll

1:21:40

again touch on this in

1:21:42

a bit, but the disk

1:21:45

photo evaporates. And then, for

1:21:47

the remainder of time, since

1:21:49

disk evaporation, Iowa, and Ganymede

1:21:51

have been... slowly migrating out

1:21:53

by tides raised on Jupiter.

1:21:56

This is the same process

1:21:58

as why the moon is

1:22:00

receding at about a centimeter

1:22:02

per year. You have to

1:22:05

enjoy it while it's there.

1:22:07

Okay, because it's it's taken

1:22:09

off, like it has had

1:22:11

enough. Okay. So the same

1:22:14

thing is happening. And naively,

1:22:16

we don't know, right, where

1:22:18

I started. Right. We know

1:22:20

that it's moving out right

1:22:23

now. We can sort of

1:22:25

do astrometry. But in fact,

1:22:27

I would argue that by

1:22:29

knowing the orbital inclinations of

1:22:31

Amaltian thebe, you can very

1:22:34

well constrain where IO started.

1:22:36

Why? Because Iyo, Europa, and

1:22:38

Ganymede all move out in

1:22:40

concert. They sweep a series

1:22:43

of interior. orbital resonances, orbital

1:22:45

resonances are configurations where the

1:22:47

gravitational perturbations between these bodies

1:22:49

become coherent, they correspond to

1:22:52

integer period ratios, and as

1:22:54

these resonances sweep, every time

1:22:56

you cross one, you get

1:22:58

a slight kick, both in

1:23:00

the eccentricity and the inclination.

1:23:03

The convergent encounters with resonances

1:23:05

lead to capture, that's how

1:23:07

i.o, Europa, and Ganymede all

1:23:09

locked into a four to

1:23:12

two to one. period ratio,

1:23:14

divergent encounters, lead to kind

1:23:16

of impulsive kicks. This has

1:23:18

been understood since at least

1:23:21

the 1980s, but probably even

1:23:23

well before that. Okay, how

1:23:25

does that work? When I

1:23:27

was a grad student first

1:23:30

working on celestial mechanics, I

1:23:32

was encountering these types of

1:23:34

diagrams, and this looks like

1:23:36

the eye of Mordor, just

1:23:38

like staring you deep into

1:23:41

your soul. But then once

1:23:43

you understand what's going on,

1:23:45

it's super clear. Okay, the

1:23:47

keys to get there. So

1:23:50

these are face space coordinates

1:23:52

and you can think of

1:23:54

the radius away from the

1:23:56

origin as the orbital inclination

1:23:59

of one of the tiny

1:24:01

satellites, say Amaltia. As Iow

1:24:03

migrates, this homoclinic curve slowly

1:24:05

contracts upon this equilibrium where

1:24:08

you sit originally at zero

1:24:10

inclination. And because this process

1:24:12

is adiabatic, face space area

1:24:14

occupied by your equilibrium is

1:24:16

conserved until you encounter the

1:24:19

separatrix. Now, the separatrix is

1:24:21

an orbit of infinite period,

1:24:23

so adiabaticity is briefly broken

1:24:25

and you acquire some face

1:24:28

space area. And then as

1:24:30

this process continues, this deforms

1:24:32

back into a circle, and

1:24:34

you have a very deterministic

1:24:37

kick in orbital inclination that

1:24:39

you can compute associated with

1:24:41

each passage of each residence.

1:24:43

So, in practice, what does

1:24:46

this mean? This means that

1:24:48

to explain a multi-ase inclination,

1:24:50

you can calculate that it

1:24:52

must have crossed the three

1:24:54

to one. orbital period ratio

1:24:57

with Iowa. If you basically

1:24:59

start Iowa too far away

1:25:01

from Jupiter, then the inclination

1:25:03

would be too small. But

1:25:06

you can't start Iowa too

1:25:08

close to Jupiter because then

1:25:10

it would sweep too many

1:25:12

residences and the inclination would

1:25:15

be too high. The same

1:25:17

argument applies to the inclination

1:25:19

of Phoebe. This is the

1:25:21

one with the 1.1. To

1:25:24

explain its inclination, you have

1:25:26

to have sweep the 6

1:25:28

to 4, 5 to 3,

1:25:30

and 4 to 2 residences

1:25:32

across this satellite. So Amaltia

1:25:35

offers a lower bound. on

1:25:37

where IO started out, namely

1:25:39

4.02 Jovian radii, and Thebe

1:25:41

provides an upper bound, which

1:25:44

is 4.06 Jovian radii. So,

1:25:46

the crater's on Amaltia. Yeah,

1:25:48

okay, great question. So yeah,

1:25:50

they're heavily cratered. They don't

1:25:53

impact because you can, so

1:25:55

they would impact rather if

1:25:57

the reoccretion time was slower

1:25:59

than the differential procession time.

1:26:01

Okay, so imagine you come

1:26:04

in, you shoot one of

1:26:06

these things, it breaks apart

1:26:08

into a bunch of pieces,

1:26:10

right? Those pieces are all

1:26:13

occupying the same orbit, but

1:26:15

those orbits can differentially process,

1:26:17

right? If the differential procession

1:26:19

takes them away, then bad.

1:26:22

but as it is, the

1:26:24

reoccretion is basically instant. Okay,

1:26:26

so by measuring the inclinations,

1:26:28

right, and matching them to

1:26:31

IO's outward migration, you can

1:26:33

constrain where I O originated

1:26:35

pretty well. I was super

1:26:37

happy when I figured this

1:26:39

out because I thought it

1:26:42

was kind of a big

1:26:44

deal, but Turns out I

1:26:46

was not the first person

1:26:48

to figure this out at

1:26:51

all. And my undergrad advisor,

1:26:53

Greg Laughlin, used to tell

1:26:55

me never fully solve a

1:26:57

problem. Like if you fully

1:27:00

solve a problem, just get

1:27:02

the full answer, then you

1:27:04

won't get cited ever because

1:27:06

there's no one left to

1:27:09

work on this problem. So

1:27:11

just get like halfway, maybe

1:27:13

70% of the way there,

1:27:15

but don't ever fully solve

1:27:17

the problem. Okay. So here

1:27:20

is an abstract. that actually

1:27:22

fully solved the problem in

1:27:24

2001 by Doug Hamilton. It

1:27:26

was never published as a

1:27:29

full paper because the abstract

1:27:31

already says everything that needs

1:27:33

to be said. Okay, it

1:27:35

basically said everything I just

1:27:38

told you. And it's got

1:27:40

a whopping three citation. Okay,

1:27:42

because they fully solved the

1:27:44

problem. And this was kind

1:27:47

of a cool discovery. And

1:27:49

by going into the Wayback

1:27:51

machine, which is like the

1:27:53

best website ever, you can

1:27:55

go and find slides from

1:27:58

a talk that Doug Hamilton

1:28:00

gave in 2001. And like,

1:28:02

there it is, right? This

1:28:04

is the inclination history of

1:28:07

Amaltia. You can see how

1:28:09

its inclination grows in this

1:28:11

step-like. fashion, very deterministic, as

1:28:13

Iow migrates out, and the

1:28:16

same is true for thebe

1:28:18

where it grows as kind

1:28:20

of a multitude of additional

1:28:22

steps. Really cool. Okay, good.

1:28:24

So now that we know

1:28:27

where Iow is, so what?

1:28:29

Well, let's go back to

1:28:31

this figure where I told

1:28:33

you early in the talk

1:28:36

that satellites and planets will

1:28:38

stop at the inner edge

1:28:40

of the disk. And in

1:28:42

fact, these types of simulations

1:28:45

have been done by everyone

1:28:47

and their brother over the

1:28:49

last 25 years, and they

1:28:51

all conclude that there exists

1:28:54

a factor of where you

1:28:56

park and where the disk

1:28:58

is truncated, and this factor

1:29:00

is close to unity but

1:29:02

slightly bigger. It's 1.13, okay?

1:29:05

The basic dynamics, by the

1:29:07

way, of what's happening here

1:29:09

is once you are... close

1:29:11

to the inner edge, then

1:29:14

you have this trailing arm

1:29:16

of the spiral density wake,

1:29:18

right? And so this is

1:29:20

a density enhancement in the

1:29:23

disk. And that's basically always

1:29:25

pulling back on the satellite.

1:29:27

And so it's sapping angular

1:29:29

momentum away from the satellite.

1:29:32

So the torque associated with

1:29:34

this arm, which is called

1:29:36

the Limblad torque, is causing

1:29:38

the satellite to go in.

1:29:40

But... Once you go close

1:29:43

to the inner edge, there's

1:29:45

also horseshoe dynamics. which is

1:29:47

you can almost see the

1:29:49

outlines of the horseshoe dynamics,

1:29:52

which is basically just taking

1:29:54

gas and throwing it into

1:29:56

the void, where it then

1:29:58

gets picked up by the

1:30:01

magnetic field and accreted. So

1:30:03

that process of throwing gas

1:30:05

in creates a torque that

1:30:07

gives the planet angular momentum,

1:30:10

and they cancel out when

1:30:12

you park the satellite. a

1:30:14

factor of 1.12, 1.13, away

1:30:16

from the inner edge. Okay?

1:30:18

So if you know where

1:30:21

Jupiter was, you can then

1:30:23

divide the... Not Jupiter, I'm

1:30:25

sorry. If you know where

1:30:27

IO was, you can divide

1:30:30

IO's primordial orbit by 1.13

1:30:32

and understand where the disk

1:30:34

was truncated. And 4.04 divided

1:30:36

by 1.3 is 3. Okay.

1:30:39

This is where... the circumjovian

1:30:41

nebula ended by the process

1:30:43

of magnetospheric truncation. Okay. You

1:30:45

had a question? Oh, it's

1:30:47

cool. Yeah, well, it's actually

1:30:50

pretty hot. Okay, right around

1:30:52

next to Jupiter, it was

1:30:54

like a thousand five hundred

1:30:56

degrees. And I have no

1:30:59

artistic skill. Okay. But, but

1:31:01

I did, I did have

1:31:03

a grant. that I could

1:31:05

do whatever I wanted with,

1:31:08

so I paid a guy

1:31:10

to draw this picture. And

1:31:12

this picture basically shows everything

1:31:14

I just said. This is

1:31:17

where the circumjobian nebula is

1:31:19

truncated, right? It's truncated by

1:31:21

the magnetic fields, you've got

1:31:23

some merdional flow, you've got

1:31:25

the thermally ionized disk, and

1:31:28

a critical consequence of this

1:31:30

truncation is that it also...

1:31:32

tells you how Jupiter was

1:31:34

rotating at this time because

1:31:37

in fact all of this

1:31:39

business with circular stellar disc

1:31:41

truncation came from the realization

1:31:43

that like t-tory stars do

1:31:46

not spin at breakup right

1:31:48

they spin at a or

1:31:50

at a period of a

1:31:52

few days and that's because

1:31:55

they spin at almost co-rotation

1:31:57

with the with the truncation

1:31:59

period of the nebula let's

1:32:01

think about how this works

1:32:03

if you write down the

1:32:06

equation for the spin angular

1:32:08

momentum of Jupiter, you've got

1:32:10

a whole bunch of terms,

1:32:12

okay, plus magnetic breaking, right?

1:32:15

This is just Lorenz torques

1:32:17

of the field coupling to

1:32:19

the disk and because the

1:32:21

disk is going Keplerian, so

1:32:24

slowly compared to say the

1:32:26

spin of the planet, the

1:32:28

field lines sap angular momentum

1:32:30

away. from the planet. You

1:32:33

also have accretion of angular

1:32:35

momentum along the magnetic field

1:32:37

lines, which is this term.

1:32:39

Now, I took some plasma

1:32:41

physics classes as a grad

1:32:44

student and my professor used

1:32:46

to tell me in plasma

1:32:48

physics you have equations with

1:32:50

lots and lots of terms

1:32:53

in them, okay, but never

1:32:55

worry because always like. two

1:32:57

of them cancel out and

1:32:59

the rest just don't matter.

1:33:02

And in fact, that's the

1:33:04

case here as well. All

1:33:06

of this first line is

1:33:08

like a 10 to the

1:33:11

minus four correction to the

1:33:13

balance of these two terms.

1:33:15

Okay, so if you were

1:33:17

to solve this, what you

1:33:19

would find quickly is that

1:33:22

J. Dot, the spin, angular

1:33:24

momentum evolution, would go to

1:33:26

zero, balanced by Lorenz torques

1:33:28

breaking the spin and accretionary

1:33:31

torque spinning up the planet.

1:33:33

and when you plug in

1:33:35

the numbers for a dipole

1:33:37

field what you get is

1:33:40

that the equilibrium rotation very

1:33:42

quickly in like 10 to

1:33:44

the three 10 to the

1:33:46

four years approaches 0.88 of

1:33:48

the orbital frequency at which

1:33:51

the disk is truncated. So

1:33:53

if you know the mass

1:33:55

of Jupiter, which I do,

1:33:57

that's 300 Earth masses, and

1:34:00

I know where Jupiter was

1:34:02

truncated, it's 3.6 Jupiter Radio,

1:34:04

I also know the period

1:34:06

with which it was spinning

1:34:09

at this time, and it

1:34:11

turns out to be about

1:34:13

a day. Now, then the

1:34:15

disk photo evaporates. Right, the

1:34:18

photovaporation front comes, reaches Jupiter,

1:34:20

and it's gone. What happens

1:34:22

after? Well, what happens after

1:34:24

is that the spin angular

1:34:26

momentum of Jupiter is conserved

1:34:29

to a great approximation. Because

1:34:31

the satellites are actually tiny,

1:34:33

compared to Jupiter, so their

1:34:35

tidal migration extracts a negligible

1:34:38

amount of angular momentum. And

1:34:40

so, if you know how

1:34:42

it was spinning, to start

1:34:44

with, and you know the

1:34:47

angular momentum now. right? You

1:34:49

know the moment of inertia

1:34:51

now and in general moments

1:34:53

of inertia can be computed

1:34:56

as a single valued function

1:34:58

of the radius with standard

1:35:00

you know planetary structure evolution

1:35:02

calculations like you know those

1:35:04

you can do with the

1:35:07

Mesa code then you can

1:35:09

just plug in the the

1:35:11

numbers and it gives you

1:35:13

what the radius of Jupiter

1:35:16

was when the disk went

1:35:18

away. It turns out to

1:35:20

be two. Jupiter was twice

1:35:22

as big as it is

1:35:25

now when the Circumjovian nebula

1:35:27

evaporated. This is a highly

1:35:29

boring answer, okay? Because before

1:35:31

I did the calculation, I

1:35:34

guessed what it was, and

1:35:36

I guessed too, because, you

1:35:38

know, people know that like

1:35:40

T- Tory stars are two

1:35:42

times the radius of the

1:35:45

sun. And I was like,

1:35:47

yeah, it's probably two Jupiter

1:35:49

radii. and this is indeed

1:35:51

a literature that people guess,

1:35:54

sorry, a number that people

1:35:56

guess in the literature already.

1:35:58

but this is kind of

1:36:00

a model independent way I

1:36:03

would argue at getting at

1:36:05

this. Okay, so what else

1:36:07

does this tell you? Well,

1:36:09

if you have the radius

1:36:11

and you have the mass

1:36:14

for a giant planet, that

1:36:16

gives you what the interior

1:36:18

entropy of the planet was.

1:36:20

And the numbers clock in

1:36:23

at a little bit higher

1:36:25

than 10 KB per barion.

1:36:27

So this corresponds to a...

1:36:29

pretty hot start of the

1:36:32

giant planet, which means that

1:36:34

most of the energy of

1:36:36

the accretionary in fall was

1:36:38

not radiated away as a

1:36:41

shock. Most of it contributed

1:36:43

to the deep interior. Okay,

1:36:45

the entropy is a subtle

1:36:47

point, it's kind of fun,

1:36:49

but let's get back to

1:36:52

something Brian said I would

1:36:54

tell you, which is the

1:36:56

magnetic field. Okay. Remember how

1:36:58

early in the talk I

1:37:01

said that for all astrophysical,

1:37:03

spherical, rapidly spinning dynamos there

1:37:05

exists a scaling law between

1:37:07

flux, like luminosity and the

1:37:10

field? You can apply that

1:37:12

same scaling law here and

1:37:14

deduce that to the extent

1:37:16

that that scaling law is

1:37:19

correct, the magnetic field of

1:37:21

Jupiter when the disk went

1:37:23

away was about 200 gals.

1:37:25

and that's a factor of

1:37:27

like 50 higher than it

1:37:30

is today. And finally, once

1:37:32

you have the field, you

1:37:34

can go back to the

1:37:36

formula of the RAM pressure

1:37:39

equals magnetic pressure to deduce

1:37:41

what the accretion rate through

1:37:43

the disk was right as

1:37:45

it went away. And that

1:37:48

gives you about one Jupiter

1:37:50

mass per million years. So

1:37:52

What do we know now?

1:37:54

Well, now we know that

1:37:57

this quasi-universal three-day pilot of

1:37:59

planets and satellites is a

1:38:01

natural consequence of the interplay

1:38:03

between disc accretion, Calvin Helmel's

1:38:05

contraction, and just dynamo generation

1:38:08

in a fully convective object.

1:38:10

And in the Jovian system,

1:38:12

specifically, you can read off.

1:38:14

what the IO initial starting

1:38:17

position was. And from this,

1:38:19

you can deduce that Jupiter

1:38:21

was twice as big as

1:38:23

it is now, when the

1:38:26

disk went away. It had

1:38:28

a field of a couple

1:38:30

hundred gals, and was a

1:38:32

creating matter at one Jupiter

1:38:34

mass per million years. But

1:38:37

like, I kept saying that

1:38:39

this is at the time

1:38:41

when the disk goes away,

1:38:43

right? This is at the

1:38:46

terminal stage of the Circumjovian

1:38:48

neb. So when is that?

1:38:50

Right? Is that one million

1:38:52

years after CAI formation? Five

1:38:55

million years? Like what's number?

1:38:57

Turns out it's 3.98. And

1:38:59

this is a well-known number

1:39:01

because of something called angrites.

1:39:04

Okay. Angrites? I always assumed

1:39:06

just stood for angry media

1:39:08

rights. Turns out it's not

1:39:10

the case. It's named after

1:39:12

some basin in Brazil, but

1:39:15

angrites are media rights. that

1:39:17

came from a parent body

1:39:19

that was volcanic. Okay, and

1:39:21

the parent body lived for,

1:39:24

you know, something like 12

1:39:26

million years. So you can

1:39:28

date them and you can

1:39:30

tell each one what time

1:39:33

after calcium aluminum inclusion formation,

1:39:35

each of these meteorites erupted.

1:39:37

But because they erupt and

1:39:39

then cooled down, they go

1:39:42

through the curie temperature. So

1:39:44

they record the magnetic field

1:39:46

that they see. And you

1:39:48

can see that at 3.98

1:39:50

million years, the field goes

1:39:53

from a couple gals to

1:39:55

like zero. And that's interpreted,

1:39:57

people in the. kind of

1:39:59

that paleomag world kind of

1:40:02

agree that what's going on

1:40:04

is they were interpreting the

1:40:06

field of the Circumsteler nebula

1:40:09

and Then once the nebula is

1:40:11