Just as drones have transformed wildlife conservation and illegal fishing patrols, they may soon make a big impression on forest conservation. Unmanned aerial vehicles could replace people in monitoring forest regeneration projects in the tropics, with consequent savings in time and money as well as much-improved data collection.
The large land area covered by rainforests makes for a traditionally labor- and cost-intensive data gathering operation, which puts it in the "too hard" category in some areas, but drone monitoring could also spread conservation efforts to new regions.
Drone-based monitoring was tested recently as part of a study by the University of Maryland and University of California, Santa Cruz, together with the Organization for Tropical Studies in Costa Rica. The researchers adopted a remote-sensing technique called Ecosynth, which generates 3D "point cloud" models of vegetation from computer vision software paired with aerial photos from run-of-the-mill 10 megapixel cameras fitted to the drones.
The US$1,500 drones captured thousands of overlapping images across 13 one hectare forest restoration sites spread across a 100 sq km (38.6 sq mi) mountainous area of southern Costa Rica.
The researchers compared the Ecosynth results with field-based human monitoring on a number of measurements – the height, bumpiness, roughness, and openness of the canopy. They also looked to see how well Ecosynth measurements predicted the abundance of fruit-eating birds, which are important to the process of forest regeneration.
Image (a) shows Ecosynth canopy height maps, while images (b), (c), and (d) show 3D point clouds for the same data, and the photos at the bottom show one of the hexacopter drones used in the study
For the most part, Ecosynth metrics were comparable to field-based measurements, although the accuracy of the drone-based method was more hit-and-miss on lower and rougher canopies. The researchers hope to rectify this (to make Ecosynth accurate in all situations) in a future study. But even now they believe that drones can begin to expand the reach and breadth of forest conservation projects.
And that, of course, could be a big win for tropical forest recovery and conservation, as many rainforest regions are difficult to access and many land owners are individual farmers who possess neither the time, the skill or the money – the existing alternative to manual monitoring, LiDAR (short for light detection and ranging), costs at least US$20,000 a flight – to monitor forest regrowth and perhaps stem the tide of mass deforestation around the world.
A paper describing the study was published in the journal Biological Conservation.