The Boring Details That Quietly Wreck Survey Data — Units, Datums, and Grid-to-Ground

Multipath and bad base stations get the dramatic case studies. But some of the most common, most avoidable survey errors come from something far less exciting: units and reference frames. In a hands-on ROAR West workshop, surveyor and geodetic engineer Rami Tamimi, founder of The Survey School, walked through a handful of these quiet mistakes that nobody notices until the numbers stop lining up.

The clearest example is a quirk unique to the United States: there are two different definitions of a foot. "There's the US survey foot and then there's the international foot," Tamimi explained, "and you don't want to know how many times I've dealt with surveys where the units were wrong." (watch at 18:02) The US survey foot converts from meters using a fraction; the international foot uses a clean decimal (0.3048). Over a short distance, the difference is invisible. "But when you start doing projections and your coordinates are in the millions, suddenly you have a shift in your data set." Six US states, including Michigan where Tamimi lives and works, still use the international foot rather than the more common survey foot — which means even surveyors working within the same country can get tripped up crossing state lines.

Tamimi gave a set of practical tells for catching a unit error before it causes real damage: a parcel's shape looks correct but its overall scale is off (watch at 19:27) (a sign that meters and feet have been swapped somewhere), a distance measurement checks out but the underlying coordinates don't, or a project in a CAD platform like Civil 3D is set to the wrong unit system entirely. Each is a symptom of the same root cause — data correctly measured, but interpreted in the wrong unit at some point in the pipeline.

A second, subtler category of error involves how a reference frame is defined. Every coordinate needs a stated horizontal datum (the X/Y reference), a vertical datum (what elevations are measured against), a projection and zone appropriate to the region, and an epoch — the specific point in time a coordinate system was fixed to, since the Earth's surface is technically always moving. (watch at 20:46) On top of that sits grid-to-ground: whether a measurement accounts for the Earth's curvature ("ground" distance, what a total station or tape measure gives you) or assumes a flat projected plane ("grid" distance, what GNSS naturally produces). "GPS thinks the world is flat," Tamimi said bluntly. (watch at 22:17) Over a short distance, the difference is negligible; over several kilometers, comparing a grid distance to a ground distance without correcting for the difference introduces exactly the kind of systematic shift that's hard to trace after the fact.

None of this requires exotic equipment or rare conditions to go wrong — it requires one unstated assumption. The fix costs nothing beyond the discipline Tamimi repeats throughout his workshop: always state your units, always state your reference frame and epoch, and always confirm whether a given distance is grid or ground before you compare it to anything else. For teams integrating data from drones, GNSS, and traditional survey equipment, this is often the actual source of a mismatch that gets blamed on "bad data" when the data itself was never wrong — it was just never told what it was being compared to.

Survey & mapping solutions · DJI drones for survey & mapping in Canada