Welcome, Guest

Author Topic: "Planetary Accretion" Test - Two weeks later, a dwarf planet emerges.  (Read 2056 times)


  • *
  • Posts: 1
Finally figured out how to do small thumbnails you can click to go to the full image! Woot!

Alright, so a few explanations before the screenshots. My only reason I bought universe sandbox was to be able to simulate protoplanetary disks and planetary accretion. Sandbox 1 wasn't up to the task, so years later, got US2. As SPH has not yet been fully implemented, we don't yet have a way to run a true protoplanetary disk and planetary accretion simulation. So I set out to try to set up something similar.

At first, I thought the easiest way would be to use the ring feature. Change particles to bodies and just fill the orbit around a star with thousands of "Random Asteroids." Turns out that no matter what I do, the bodies always spawn as Earth Moons rather than bodies of smaller sizes. No clue why. Apparently no option to choose what kind of body to spawn too? Maybe I'm looking in the wrong place. Let me know below if I am wrong. Also, I found that having a star at the center of the system, due to the lighting, was really GPU/CPU intensive. The total number of bodies I could place before the time step went near zero was far too small to be very useful.

So, I decided to do a "moon accretion" sim instead, using Jupiter as the "star" of the system and placing 900-950 Random Asteroids in orbit around Jupiter, then adding more each time it got down around 300-400 objects from collisions. I've been doing this constantly, for two weeks (~460 years ingame time), adding ~900 Random Asteroids, letting it run, then adding more. All Random Asteroids have been placed in plane with Jupiter in circular orbits to try to keep it a bit tidy, but several asteroids due to gravitational interactions now have very peculiar orbits.

It's worth remembering that the Random Asteroid tool places what essentially look like tiny rocks. They are not rounded at all and are much smaller than Moons or even Small Moons. Looking at the range of masses on the Random Asteroids in the system at the moment, the smallest is 3.69 x 1016 kg. For perspective, this is about 1/25,700 the mass of Ceres. Massive, but tiny in US2.

As these tiny asteroids collide with others and grow, their shape rounds as they become more massive just like in real life, but US2 also does us the service of beginning to improve the texture placed on the object once it reaches decent mass limits. Not surprisingly, these texture improvements begin when asteroids begin to get massive enough to be considered dwarf planets based on their spherical shapes. These rounded, "proto" dwarf planets started appearing around day 4 maybe. It was clear they were different from the asteroids because they were rounded, but they still lacked a unique texture based on their materials percentages. Here are some current "proto" dwarf planets currently still in the simulation.

On the left, you can see the beginnings of dwarf planets. On the right, you can see larger asteroids which are not yet massive enough to be forced into a sphere by their own gravity. Based on the current division in the models in the sim right now, it looks like US2 makes this divide somewhere around 9.45 x 1018 kg. In the above screenshot, I have cut off the two most massive objects in orbit around Jupiter, because those are the stars of our post. Behold, Rnax and Ictonaer!

At 1.57 x 1021 kg, Rnax surpasses the mass of Ceres at 9.47 x 1020 kg. However, it is still only about 2% the mass of Earth's Moon. And yes, that is liquid standing water on the surface of Rnax. I'll explain below.

Ictonaer is less massive than Rnax at a mere 7.44 x 1020 kg. In fact, it is only about 78% the mass of Ceres, but has been given its own unique texture, so I'm betting US2 acknowledges it as a dwarf planet sized body :)

One of the interesting things about the Random Asteroid tool is that the random asteroids it spawns have various material percentages. As such, it is possible that certain dwarf planets that form have predominantly iron, some more water than others, etc. In our particular simulation, Rnax currently has liquid water on its surface. At about 4.9% water by mass, it's actually a pretty watery place. However, due to our simulation not being around a star, there is no habitable zone. Jupiter does not produce heat, and I believe I have tidal heating turned off (is that feature even working yet without SPH?) So how does Rnax have water sitting on its surface with a balmy 160 degree surface temperature while Ictonaer (which also has a decent amount of water) lay frozen? I had to run the sim a while and watch it to find out.

It ends up that Rnax, being the most massive object in the sim other than Jupiter, is more likely to get collisions with small asteroids than Ictonaer. Small asteroids don't heat up the dwarf planets much, but Rnax is big enough to get hit by enough of them that it often changes from a snowball to having visible liquid water. Ictonear, although it does occasionally get impacts, doesn't get them often enough to stay unfrozen for long if at all. If I were to stop adding new asteroids, I'm sure eventually Rnax would cool down and return to its former iceball self.

To give you an idea of just how many small asteroids there are compared to the small number of proto dwarf planets and dwarf planets, here is the full chart of system bodies.

By my count, out of 949 orbiting objects, 19 count as proto dwarf planets, 2 as true dwarf planets, and currently 0 as true planets, as none of them have completely cleared their orbit or similarly sized objects, although Rnax is coming close to being able to make that claim.

This is a top down screenshot of the orbits of the 21 most massive orbiting objects. As you might be able to guess, Rnax is the innermost orbit, being the most massive it has done a pretty good job of clearing its orbit.

This is a top down screenshot of the orbits of only the 6 most massive orbiting bodies. I chose 6 as the cutoff arbitrarily, but there was a large and mass size difference between 6th and 7th bodies, so I felt it was as good a divide as any. In order of mass, Rnax, Ictonaer, Oman, Manc, Cimanasci, and Oenti. I hypothesis that due to the limitations of US2 and my computer, plus the larger relative space available in outer orbits, I haven't been able to place enough asteroids in outer orbits to get collisions at the same rate as the more crowded inner orbits. This leads to a general decrease in mass per object the further out you go. It's worth remembering that I did not choose any of the names for these bodies, as US2 provides a randomly generated name for each asteroid upon creation.

As of writing this post, I have at no time observed a proto dwarf planet or dwarf planet catching a small asteroid in its own gravity to even temporarily claim its own small captured moon. However, tests in other simulations with orbiting bodies of more drastic mass differences showed me that US2 is capable of allowing an orbiting body to capture another body orbiting the same parent body, as the gas giants in our solar system capture sun-orbiting asteroids and claim them as satellites. I hope to one day see that in this sim. Rnax is lonely :)

I have to say, this sim has been an emotional roller coaster of sorts. There were several other massive bodies I was rooting for over the past two weeks, but were eventually swallowed up by Rnax... Interestingly, Ictonaer was not in the original lineup of more massive bodies. I don't remember seeing it until a week in- an underdog of sorts that must have gathered enough mass while I was sleeping one night to absorb another massive body, thus getting itself a spot in the top five.

If anyone else here runs sims similar to this one, please tell me your stories, post screenshots, etc! I realize many of us want protoplanetary disks and we're not entirely satisfied with the random solar system button, as this is mainly about the journey and not the destination. So let's live vicariously via each other's stories while we wait for SPH!
« Last Edit: March 17, 2016, 03:05:42 PM by Megneous »

Gordon Freeman

  • *****
  • Posts: 480
Re: "Planetary Accretion" Test - Two weeks later, a dwarf planet emerges.
« Reply #1 on: March 18, 2016, 10:10:14 AM »
This is the most impressive thing I've seen anyone do with US2. Awesome job!


  • ****
  • Posts: 38
Re: "Planetary Accretion" Test - Two weeks later, a dwarf planet emerges.
« Reply #2 on: March 19, 2016, 01:25:51 PM »
I might just have to try this.  when you add the asteroids are you just clicking like mad or is there a better way?


  • ****
  • Posts: 38
Re: "Planetary Accretion" Test - Two weeks later, a dwarf planet emerges.
« Reply #3 on: March 20, 2016, 08:37:43 PM »
Sooo turns out that once I get enough objects to start lagging my CPU runs at 60+ C.  I don't want to do that for long periods... so until I get an upgrade I'm afraid it won't work for me.