1,000-ha catchment, 4 villages, client's own planting records: 82% of area confirmed at harvest window; a red-rot-pattern stress cluster flagged in a single village — weeks before scouting. Flags re-verified 4/4 in our regression harness.
Species/age/area inventory + harvest verification across 4 villages. All 68 detected harvest events reproduce exactly under re-audit; a 12-day early-warning tier now fires a median 36 days sooner.
Real declined-pixel scout pins (not a sampling grid) + dated EUDR screen against the Dec-2020 cutoff. 17/20 pins persist on live re-scan; the 3 marginal ones carry "verify-first" priority — that nuance is on our validation page, not hidden.
Live plucking-season scan on the estate's own coordinates found a previously unflagged weak block — re-confirmed on fresh data five weeks later. Meanwhile monsoon cloud blinded optical entirely at those blocks for 8+ straight weeks: the exact gap radar closes.
Farm-specific sea-surface temperature at 4-km resolution vs the 25-km regional blur (up to 2.1°C apart at the farm), plus a modern multi-sensor chlorophyll layer with disclosed cross-product uncertainty.
Plot-by-plot screening against the regulation's own cutoff — 11,500-plot inventory screened (0% post-2020 risk), by-year evidence, audit-ready lineage. Supports, not replaces, traceability due diligence — we say so.
Radar imaged one of Ontario's largest-ever fires at 98% coverage on ignition day; optical needed 12 days to see even half the burn zone. Measured this season, FIRMS ground truth — including the honest negative on what C-band can't map in boreal fuels.
Open the fire case →A real Hoeksche Waard arable field: healthy crop, uniform cover, and the all-weather story — only 30% of optical passes clear in a year vs 76 radar passes. Hand us a boundary at the meeting and get this on your own ground by coffee.
Open the field scan →A real Saskatchewan grain field: crop peak in August, harvest dated to the September pass, on a uniform field — the same read runs across thousands of quarter-sections at once. Honestly framed for the dry Prairies: scale + lodging/hail/claims, not an all-weather-frequency story that doesn't hold there.
Open the broadacre case →A real managed pine plantation (public forest, no client): 20 canopy-change hotspots detected against a 2019–2023 baseline — clearfells, thinning and any unplanned loss — each a clickable pin your crew can open in Google Maps and drive to. The change map catches your planned fells (proof it works) and the losses you didn't plan (the value).
Open the interactive scout map →