# Universe Sandbox 2 accurate representation of gravity

I've been playing Universe Sandbox for a while now and I have never been able to make a near Earth asteroid collide with it. By "make" I mean, simulating billions of years of known, near Earth objects until one of their orbits intersect. In this explanation Sir Cumference describes that Universe Sandbox 2 uses Newtonian physics to in its simulations. But wouldn't that calculate enough pull from Earth to slightly change the trajectory of the NEO to eventually impact?

We know, with almost 100% certainty that a LARGE NEO will impact Earth (probably sooner than later), otherwise why would we be investing so heavily on the Jet Propulsion Lab's Asteroid Watch program.

Can the game not accurately simulate gravitational pull on those large scales or is there something else (probably a lot) I'm missing? Are all known NEOs not projected to have a relevant collision intersection with Earth?

A NEO only impacts Earth if both bodies are in the orbit intersection zone at the same time, closely enough for gravity $(F \propto 1/r^2)$ to bring them in contact. In a simulation, the timestep should also be short enough to detect a collision. An object at a moderate relative speed of 10 km/s crosses Earth's diameter in about 20 minutes; a longer timestep could turn a hit into a miss.

If you're already modeling real asteroids, try some which have hit Earth: 2008 TC3, 2014 AA, 2018 LA. The short observation arcs make their pre-impact orbits highly uncertain, so a few small bodies spread within $\pm\sigma$ of each orbital element value may improve your chances. Alternatively you could set an object up for immediate collision with Earth at a relative speed of 12 to 20 km/s in any direction, and run the simulator backward to see how it would get there.

The asteroid risk is real, but we should not overstate it. If you visit JPL CNEOS Sentry and use unconstrained settings, it shows only a few objects with more than 1 chance in 1000 to hit Earth in the next 100 years, and most of those are small. Few NEO orbits are known precisely enough to make meaningful predictions beyond that. The Palermo scale quantifies a given object's threat to Earth in the foreseeable future.

• Gravitational attraction can certainly help, but it's not required. Geometry alone can do it.
– uhoh
Jun 29 '18 at 8:17
• So I have since re-ran the simulator on shorter timesteps when the NEO comes closer to my simulated Earth. I had to create a new asteroid that came ridiculously close and after a bunch of unsuccessful attempts I finally got one try close enough for Earth to pull its orbital path on a collision course. I ended up almost sychronizing my new body's year to one Earth year to get there faster. Thanks for the links. Jun 29 '18 at 21:20
• @FusRoDah I updated the known impactor links to show orbit diagrams. The asteroid orbit need not match Earth's as long as they cross and the timing is right. Jun 30 '18 at 17:18

Impact probabilities are probabilities because there are uncertainties in the orbits themselves, as well as in the simulation or propagation.

In other words the measurements used to produce the starting points have plenty of observational uncertainties, and the error in the simulation grows due to these uncertainties grows the longer you run it for.

So even if you had a "perfect" simulator, you'd have to run the simulation kerjillions of times to include a thousand slightly different staring point for each of the bodies in the simulation. Since everything interacts with everything (in the case of gravity) that becomes a huge problem. Real calculations use all kinds of techniques to try to include uncertainties in their propagator, hopefully another answer here will expand on that.

Then there are uncertainties in the non-gravitational forces such as outgassing and pressure from sunlight and the solar wind.

To read more on that, see the new BBC News article Interstellar visitor's identity solved and also scroll down to the link that says The full study is published in Nature where an open access version of the Nature paper Non-gravitational acceleration in the trajectory of 1I/2017 U1 (‘Oumuamua) is available for you to read.

• First learned "kerjillions" from Laurie Anderson's Ramon "So when you see a man who's broken, Pick him up and carry him, And when you see a woman who's broken, Put her all into your arms, 'Cause we don't know where we come from, We don't know what we are..."
– uhoh
Jun 28 '18 at 23:10
• +1 for kerjillions. But seriously thanks for the reads Jun 29 '18 at 21:14