# What does it mean to ask for a date beyond epoch time?

I am fairly new to this industry. As I am playing around with a library on Python, ephem, I have encountered an error saying that:

TLE elements are valid for a few weeks around their epoch, but you are asking about a date X days from the epoch

My first question is that what does it mean? Can I manually fix it?

• I've added the pyephem tag, and the skyfield tag as well, since pyephem is now in the process of being being depreciated, and (as far as I know) both use ephem also supported by the same person. I've added the sgp4 tag since that's what propagates TLEs.
– uhoh
Jan 17, 2019 at 19:28

A TLE describes the orbit of a body at a single reference point in time, which is called the epoch. It is straightforward to extrapolate the position of the object for some time forward or backward from that snapshot, but beyond that, limitations of the propagation model cause errors to grow, and the position of the object cannot be accurately estimated.

The error message you're getting is telling you that you're using an outdated TLE. The solution (assuming you're trying to find the present location of an object) is to find a newer TLE for the object. How you do that depends on what the object is, and how you got the old TLE in the first place. It may be as simple as updating to a more recent version of pyephem.

If you’re asking for a projection of an Earth-orbiting satellite far into the future, pyephem may not be up to the task.

Not necessarily "outdated" as much as simply exactly what the ever-articulate (see also) Brandon Rhodes has written: "X days from the epoch" (unsigned, could be either direction) because TLEs are sometimes post-dated to the future by as much as several weeks, and PyEphem may feel that it is too early to use that TLE as well as too late.

note: pyephem (deprecated, "old-and-busted" (not really)) and skyfield ("new hotness") are supported by the same person.

Don't believe me? Then see the excellent answers to the question Can the TLE epoch be listed as in the future? How?

See @Tristan's excellent and thorough answer to the question Is SGP4 propagation necessarily more accurate near the epoch chosen for TLE generation?

It also may not be at all that "pyephem may not be up to the task." as much as it's simply recognized that any use of SGP4 (pyephem or otherwise) is known to give increasingly worse results away from epoch. This would be true of anything accepting a TLE because it would be implementing SGP4 (or worse). See all these linked answers for this sites best explanations of this (in my opinion).