Tropical rainforests worldwide are negatively impacted from a variety of human-caused threats. Unfortunately, our ability to study these rainforests is impeded by logistical problems such as their physical inaccessibility, expensive aerial imagery, and/or coarse satellite data. One solution is the use of low-cost, Unmanned Aerial Vehicles (UAV), commonly referred to as drones. Drones are now widely recognized as a tool for ecology, environmental science, and conservation, collecting imagery that is superior to satellite data in resolution. We asked: Can we take advantage of the sub-meter, high-resolution imagery to detect specific tree species or groups, and use these data as indicators of rainforest functional traits and characteristics?
We demonstrate a low-cost method for obtaining high-resolution aerial imagery in a rainforest of Belize using a drone over three sites in two rainforest protected areas. We built a workflow that uses Structure from Motion (SfM) on the drone images to create a large orthomosaic and a Deep Convolutional Neural Networks (CNN) to classify indicator tree species. We selected: 1) Cohune Palm (Attalea cohune) as they are indicative of past disturbance and current soil condition; and, 2) the dry-season deciduous tree group since deciduousness is an important ecological factor of rainforest structure and function.
Results/Conclusions
Preliminary results show distinct patterns in the distribution of both Cohune Palm and deciduous trees. The image resolution was 5 cm/pixel (572-hectare site - BFREE), and 10 cm/pixel (445 ha and 462 ha sites - BNR). Using YOLOv2 CNN to classify the individual trees resulted in an average precision (AP) of 79.5% for Cohune Palm and 67.3% AP for deciduous trees. We identified 6,308 palms at BFREE, yielding 120.4 ha palm coverage. Palms were more prevalent at BFREE, (disturbed) with 21% palm cover, compared to only 5.9% and 2.0% in our less-disturbed, mountain sites (BNR). Total palm coverage in the BNR ranged from 6.2 ha - 27.3 ha. Our preliminary data for deciduous trees at BFREE resulted in identifying 2,389 trees with a total coverage of 17.8 ha or 3.0%. Our results conform well with past on-the-ground data for both tree types. Cohune palms tend to grow in disturbed areas, but also contribute to soil organics. Deciduous tree crown area calculated in comparable rainforests ranged from ~3-10%. This study shows how UAV and CNN can save time (vs. manual) identifying specific tree species, helping to determine key rainforest habitat characteristics, as well as aid research and management of remote areas.