Reference Frame and Datum Mistakes
The second companion in this series explained what a reference frame actually is, and why more than one valid realization of one exists at once. The word datum, often used almost interchangeably with reference frame in everyday field conversation, gets its own careful treatment here for a narrower reason: this chapter asks a practical question neither prior companion focused on directly: how would a mismatch, in either the frame or the datum built on top of it, actually be caught, before it quietly ends up in a deliverable?
A reference frame or datum mismatch produces a coordinate that is perfectly precise and completely self-consistent, differing from the intended answer by an amount that can range from a few centimeters to well over a meter, with nothing in the FIX status to reveal it.
Why This Error Hides So Well
A rover tied to the wrong reference frame, or converted through the wrong datum, will still resolve FIX cleanly, still repeat its own answer precisely on reoccupation, and still look completely healthy by every internal measure the receiver can report. The mismatch lives entirely in a piece of metadata, which frame, which epoch, which datum, that the receiver has no way to question on its own.
What Actually Catches It
Comparing a fresh coordinate against a nearby point with a known, published value in a trusted reference frame is the most direct check available, since a mismatch typically shows up as a discrepancy that is far too large to be ordinary RTK noise, yet still oddly consistent across an entire project. Confirming, explicitly and in writing, which reference frame and epoch a base coordinate or a correction service is providing, rather than assuming it matches the project's intended system, closes the gap before it ever becomes a field discrepancy.
A Question Worth Asking Before the Survey Starts
The most efficient moment to catch a frame or datum mismatch is before fieldwork begins, by simply confirming, out loud and in writing, which frame, which datum, and which epoch a project is standardized on, and verifying that every correction source and base coordinate in use actually matches. Catching this after the fact requires comparing against independent control. Catching it beforehand requires only a conversation.
Every point on a site is offset from expected control by nearly the same amount, in nearly the same direction, regardless of which part of the site was surveyed or what the satellite geometry looked like that day. What category of error does that pattern point toward? A shared, systematic issue, most likely a reference frame, datum, or base coordinate mismatch, rather than anything related to individual observation quality.
A reference frame or datum mismatch produces a precise, internally consistent, and completely wrong coordinate, with no warning visible in the FIX status. It is caught by comparison against independent, trusted control, or better yet, prevented entirely by explicitly confirming frame, datum, and epoch before a survey begins.
Network RTK introduces its own, more specific quality considerations on top of everything covered so far, and the next chapter looks at exactly what those are.