Grassland and livestock

On this page 3

Silage, grazing and the wildlife obligation that comes with a mower.

Grassland and silage

Search before the mower moves

Thermal fawn searches ahead of first cut, sward quality mapping, and reseeding where the ground has gone back.

The challenge. First cut silage is the single most lethal operation of the farming year for wildlife. Roe deer leave fawns lying in tall grass and a fawn will press flat rather than run from a mower. In Germany the duty to take reasonable precautions falls on whoever is doing the mowing, and a line of volunteers walking a 12 hectare field before dawn is neither reliable nor repeatable. Separately, sward quality drifts: patches go back to weed grasses and nobody maps it until the analysis comes back poor.

The approach. A thermal search flown in the cool hours before the mower starts, with finds pushed live to the phone of whoever is on the ground, and a coverage record kept per field. Sward composition is mapped between cuts so reseeding is targeted. Drone broadcast handles the wet hollows and steep banks a machine should not be on.

WindowFocus
Before first cutThermal wildlife search with live find marking and a coverage record
Between cutsSward quality and weed grass mapping
Later cutsRepeat searches on fields with known activity, plus yield estimates per cut
Late summerOverseeding of thin areas by drone where machinery cannot travel
AutumnDrainage and wet area mapping for the winter works list

Capabilities: wildlife, field mapping, seeding, yield, livestock.

Outcomes. A search record per field, signed off before the mower starts. Finds logged with coordinates and outcome. Reseeding targeted at mapped thin areas. Cut yields estimated rather than counted off the clamp.

Scale. Fields of 2 to 40 hectares, flown at dawn.

Why this is the best first market. The window is two hours before sunrise, the payload is one thermal camera, the pipeline is frames in and points out, the legal footing is clear, and the customer knows within an hour whether it worked. See the build.


Dairy and grazing livestock

The outlying block, checked every morning

Daily head counts, fence line inspection and flagged animals, without the drive to every outlying block.

The challenge. The blocks furthest from the yard get checked least, and they are the ones where a fence going down matters most. A daily round of outlying grazing is an hour of driving that finds nothing on most days. On the days it would have found something, the delay between the event and the check is the whole cost: an animal through a fence onto a road, or a cow down in a corner since the previous evening.

The approach. A scheduled patrol per block, flown from a dock rather than driven to. Head count against expected, fence line imagery, and thermal signatures that separate an animal lying and ruminating from an animal lying and not. Anything ambiguous goes to a person before it becomes an alert.

WindowFocus
TurnoutFence line survey of every block before stock go out
Grazing seasonDaily patrols with head counts and flagged animal reports
Peak growthGrass cover mapping for the grazing rotation
AutumnPoaching and damage mapping for the reseeding plan
HousingFence and water point condition survey for the winter works list

Capabilities: livestock, herding, wildlife, field mapping, network, seeding.

Outcomes. Head count per block recorded daily against expected. Fence faults reported the morning they appear. Flagged animals delivered with imagery for a human decision. Grazing rotation planned from mapped cover.

Scale. Grazing platforms of 20 to 300 hectares, docks placed by battery range.

Build note. This is the use case that justifies a dock, because its whole value is frequency without a drive. It is also the use case that most needs the regulatory work done properly: a daily patrol over grazing is usually beyond visual line of sight. Start with the blocks the pilot can see from the yard, prove the product, then do the authorisation.

One aircraft, two seasons

The thermal payload that searches for fawns in May patrols the herd in February. Same airframe, same camera, same time of day, opposite ends of the calendar. For a livestock farm or a contractor, that is one purchase covering both peaks, which is the single best hardware economics in the whole platform.