200 ha LiDAR survey of the Okpara Farm

The context

Société Nationale de Mécanisation Agricole (SoNaMA) operates the Okpara integrated agricultural site, a 200-hectare zone near Parakou. To plan irrigation, monitor plot evolution and identify trees exceeding plant-health thresholds, they needed centimetric multi-layer mapping: bare ground under vegetation, canopy, water bodies and tree inventory. Classical aerial photogrammetry alone would not have reconstructed the ground under the canopy. Airborne LiDAR offered the complete solution in a single mission.

Our method

Three days of high-density airborne LiDAR capture, paired with a photogrammetric RGB flight for colour. Layout of RTK-georeferenced GCPs across the entire site to guarantee centimetric absolute accuracy. Multi-layer processing: DTM extraction via automatic ground classification, CHM generation by DSM-DTM subtraction, automatic detection of each tree via canopy segmentation. Manual QC complemented the automatic classification on areas where the LiDAR signature was ambiguous.

What we delivered

• Georeferenced centimetric orthophoto (GeoTIFF, GSD 2 cm/pixel)
• DTM and DSM in GeoTIFF format, UTM 31N projection
• CHM (Canopy Height Model)
• Classified LAS/LAZ point cloud (ground, vegetation, built, water)
• GIS layer of detected trees with individual height (SHP, GeoJSON)
• Thematic mapping of trees exceeding 17 metres
• Topographic plan usable for irrigation planning
• Technical report with QC and achieved accuracy

The outcome

SoNaMA now has a georeferenced cartographic layer directly usable in their internal GIS. Automatic detection surfaced specimens above thresholds that had been previously uninventoried. The data also serves as a temporal reference: a second pass in 12 or 24 months will precisely measure canopy growth and plot evolution. The mission demonstrated that an airborne LiDAR survey on 200 hectares of integrated agricultural zone is delivered in under three weeks, vs several months for an equivalent manual inventory.