Tree inventory and invasive species survey

The register that is always out of date

A tree register is accurate on the day it is finished. After that it drifts: a tree comes down in a storm, a planted whip fails, an ash declines, and nobody updates the spreadsheet. Invasive species are worse, because they are recorded from wherever someone happened to be standing when they spotted them.

Land managers have the knowledge. What they lack is a complete, current picture of the ground to hang it on, and the hours to walk every metre of it to build one.

Every tree, counted and placed

Terravect’s models detect each tree in a survey and mark it in place on a 3D model of the site, so the count and the map come from the same flight. Each tree carries a position, a height and a crown spread, read from the survey rather than estimated from the path.

That gives you:

• A register that starts complete, rather than one built tree by tree over several years.
• A baseline to repeat against. Fly the site again next year and you can see what has gone: lost trees, failed planting, new self-seeding.
• Planting you can account for. Count the saplings that took, by block, instead of walking the lines.
• A shortlist for the ground. Crowns that are thinning or dying back stand out from above, so the walk starts at the trees that need it. A survey does not replace the on-foot inspection; it tells the tree officer which trees to inspect first.

Sorting what you find

A count says how many. Classification says what. Terravect groups detected trees by type, so a site can be read as broadleaf and conifer, as mature and young, as healthy and declining, not as a single undifferentiated canopy.

This is where the work is. Ash dieback, for instance, shows from above as a crown that is thinning and dying back among crowns that are not, and the pattern across a site tells you whether you have a scatter of cases or an outbreak. A photograph taken from the path cannot show that.

Knotweed and the other plants nobody wants

Japanese knotweed is the one everyone knows, but a conservation charity or a council parks team is usually watching several at once: Himalayan balsam along a watercourse, giant hogweed on a verge, rhododendron under a woodland canopy. All of them spread, and all of them are cheaper to deal with the earlier they are found.

A drone is good at exactly the part people find hardest: the whole site at once, including the bankside, the back of the allotments and the strip behind the depot that nobody walks. Knotweed stands are tall, dense and a distinctive colour for much of the year, and they stand out against grass and scrub in a multispectral survey and in the height model. Terravect marks each stand where it sits on the 3D model, with an area measured from the survey.

There is a difference between what the survey shows and what is under it. Knotweed spreads below ground, and a survey sees the stand, not the rhizome. A plant growing under dense tree canopy is also harder to see than one in the open. Both mean the survey is where the work starts and the specialist contractor does the rest. But it starts from a map of every stand on the site, not from the three someone remembered.

Mapping before anyone digs matters for another reason too: soil and plant material from a knotweed stand is controlled waste, and moving it by accident is expensive.

Many people, one model

Conservation work and public land are done by many hands: a ranger, a tree officer, a contractor, a volunteer group, a landowner next door. They mostly share information the hard way: spreadsheets, emailed photos, “the one by the bench”.

On Terravect they work on the same model of the site. The ranger drops a note on a stand of balsam. The contractor sees it with the photograph and the measured area. The volunteer group adds what they pulled on Saturday. The tree officer sees the crown that has been flagged and what was said about it, in the place it is. Nobody has to ask where “the one by the bench” is.

The next survey then shows what actually happened: whether the treated stand has regrown, whether the planting took, whether the balsam came back on the far bank.