The Earth Moves, So Coordinates Have Dates
A monument can sit undisturbed for decades, its physical mark untouched, and still be attached to a coordinate that no longer matches what a receiver reports today. Nothing about the monument failed. The ground it is fastened to has been quietly moving the entire time.
Tectonic plates move continuously, often a few centimeters a year. At the accuracy RTK provides, that motion is not negligible, and a coordinate is not complete until it states the moment in time it describes.
A Few Centimeters a Year Adds Up
A few centimeters of annual drift sounds trivial next to the scale of a continent, but RTK routinely resolves positions to a centimeter or better. Over ten or fifteen years, plate motion alone can shift a point's coordinates by a distance well within what a rover can actually detect. A survey from a decade ago and a survey performed this afternoon can both be entirely correct and still disagree, for a reason that has nothing to do with either instrument.
Coordinate precision and coordinate age are independent properties. A perfectly precise position from ten years ago can still disagree with an equally precise position taken today, because the reference frame beneath both of them has continued to move.
Two Dates, Not One
Modern high-precision work distinguishes between two different moments. The observation epoch is simply when the antenna actually collected data in the field. The reference epoch is the date a coordinate has been expressed at, or transformed to, often to allow a fresh measurement to be compared fairly against older project data. A position observed this year can be reported at this year's observation epoch, or shifted mathematically to an earlier reference epoch to match a long-running project, using a known velocity for that patch of ground. Both descriptions can be correct. They are simply answering different questions: where was this point measured, and when should its coordinate be understood to apply.
A rover occupies the same monument this year that a crew surveyed a decade ago, and the new coordinate differs slightly from the old one. Does that mean either survey was wrong? Not necessarily. It may simply mean the ground moved, and neither coordinate was expressed at the other's epoch.
The Coordinate That Finally Has Everything It Needs
Put the last several chapters together and a solved GNSS position now carries everything this companion has established so far: a raw three-dimensional location, expressed in a specific reference frame, tied to a specific moment in time. That is already a complete, physically meaningful description of a point. It is also still not the number most people expect to see on a screen.
A coordinate is not complete without a date attached to it, because the ground beneath every monument moves continuously. The observation epoch and the reference epoch answer two different questions, and a meaningful comparison between two surveys requires knowing both.
A position, a frame, and an epoch are still expressed as raw three-dimensional distances from the Earth's center of mass. The next chapter finally makes the translation most readers have been waiting for, from that raw description into the latitude, longitude, and height a person can actually read.