Does current Science technology measure the orbit of planets accurately?

Question

In past, people tried very hard to measure the orbit of planets based on observation, math and geometry without much accuracy. Nowadays, can our Science technology measure the orbit of planets accurately enough for a long period? The orbit of planets are affected by other planets with different weighting in 3-D, and there are many unknown planets in Universe, which cause any unknown gravity forces from a unknown distance. With current Science technology, do we simply predict the orbit of planets within a limited period [such as 100 years] by fitting an existing orbit of each planet? [Curve Fit] If any error occurs, then do we simply make some adjustment on existing curve of orbit? I would like to know what kinds of Science Technology we currently have to track the orbit of planets accurately. Do we have a true formula to measure the orbit of planets accurately for the next million years? Could anyone tell me how to reply someone's answer for more interpretions on my post? Thank you

Answer 1

The Jet Propulsion Laboratory (JPL) has a "long" ephemeris that covers the years 3000 BC to 3000 AD, and a regular ephemeris for the years 1600 to 2200.

The accuracy is *very* good. Take Mars, for example: we can back-predict the year 1600 position within 20 kilometers, and predict the year 2200 position within 8 kilometers. For the current years, the accuracy is better than 1 km for the 4 inner planets.

The main reason for errors is the gravity perturbations of many asteroids whose masses are not well known.

For orbits over millions of years, we can predict the shape of the orbit very well, but not at all where the planet is on the orbit.

Answer 2

There are a couple of ways to get the accuracy you are talking about.
1. Come up with a perfect model of the solar system (all the way from the sun, to past the outer edges of the Oort cloud; including everyting from the largest entity (sun) down to the smallest micrometeorite. Then solve the billion or so particle model for the complex gravitational field.

2. Take accurate observations on the largest bodies and perform a numerical analysis, solving for position, velocity, etc.

Doing #1 is not possible, because we don't even know 1/2 of what's really out there (especially in the outer Oort cloud), and there will always be unpredictable (i.e. rogue bodies) things happen to the model.

#2 is how 'they' do it. I don't think the models used would work to predict a planet's position for a million years, but they are certainly good enough for the next 25 or so years. Right now there is a database of positions based on observations calculated for the year 2000. This database is updated regularly, and perhaps another answerer can tell you how often (10 years, 20 years?)

.

Answer 3

Yes. The positions of the planets are currently known with great accuracy, and their positions have been predicted far into the future. The first person to make truly accurate measurements of their position was Tycho Brahe in the 16th century, and his positions were sufficiently accurate for Kepler to discovers his laws of planetary motion. Quite amazingly, Tycho's observations were made without any optical aid, as the telescope hadn't been invented, but he built large wall quadrants to make his measurements. Royal Greenwich Observatory in England started making accurate transit observations in the 17th century, and these continue to this day at national observatories around the world. This was mainly done for timekeeping, but that is now done with atomic clocks. Navigation of space probes, for example, would be impossible if the positions of the planets and their moons weren't accurately known. Radar is currently used to determine accurate positions for planets and asteroids, including the reflectors placed on the Moon by Apollo astronauts.