Drone surveys on a 3D digital twin

What a digital twin actually is

A digital twin of a farm, ground or course is a georeferenced three-dimensional model of the real ground — its shape, its surfaces and its features — that survey data is laid over and updated against as new passes come in.

The word that does the work is updated. A one-off 3D model is a model. A twin is a model that keeps pace with the thing it represents, so that every new survey lands in the same frame of reference as the last one.

How it differs from an orthomosaic

An orthomosaic is the standard deliverable from a drone survey: individual frames stitched and geometrically corrected into one large, flat, top-down image with consistent scale. It is genuinely useful, and it is also flat by construction — the correction that makes it measurable is precisely what removes the relief.

That removal costs you three things:

• Slope and aspect. A south-facing bank dries faster than the flat ground twenty metres away. On a flat map those two areas are neighbours with no explanation for their difference.
• Drainage. Water moves according to the shape of the ground. Where it collects and where it runs off predicts a good deal of what a moisture-sensitive index will show.
• Vertical structure. Canopy height, tree lines and shelterbelts influence the ground beneath and around them, including shading that a top-down view reads as a low-vigour patch.

Terrain as explanation

This is the practical argument for the twin. A great many of the patterns in a vigour map are not agronomic mysteries at all — they are the terrain showing through. A dry strip along a ridge, a persistently strong patch in a hollow, a band of weak growth in the shade of a tree line.

Seeing a finding in place, on the shape of the ground it came from, sorts the patterns that need investigating from the ones that simply reflect where the land sits. That is the difference between a map that raises questions and one that answers them.

Change over time

A survey is a snapshot. The value compounds when snapshots stack up and you can see movement — a patch that is spreading, a stressed zone that recovered after irrigation, a problem that reappears in the same corner every season.

A shared frame of reference is what makes that possible. When each pass is laid over the same model of the same ground, the comparison is like for like: you are looking at the field changing, not at two flights that happened to be flown differently.

Finding the thing on the ground

There is an unglamorous, entirely practical benefit too. Somebody has to walk out and look at whatever the survey flagged, and a finding placed on a recognisable model of the land — next to the track, the gate, the third green, the sixth row of that block — is a finding that gets found. A coordinate on a colour ramp is not.