Search for: All records

Abstract Pinniped species undergo uniquely amphibious life histories that make them valuable subjects for many domains of research. Pinniped research has often progressed hand‐in‐hand with technological frontiers of wildlife biology, and drones represent a leap forward for methods of aerial remote sensing, enabling data collection, and integration at new scales of biological importance. Drone methods and data types provide four key opportunities for wildlife surveillance that are already advancing pinniped research and management: 1) repeat and on‐demand surveillance, 2) high‐resolution coverage at large extents, 3) morphometric photogrammetry, and 4) computer vision and deep learning applications. Drone methods for pinniped research represent early stages of technological adoption and can reshape the field as they scale towards the full potential of their techniques. 

Abstract Human activities and climate change threaten seabirds globally, and many species are declining from already small breeding populations. Monitoring of breeding colonies can identify population trends and important conservation concerns, but it is a persistent challenge to achieve adequate coverage of remote and sensitive breeding sites. Southern giant petrels (Macronectes giganteus) exemplify this challenge: as polar, pelagic marine predators they are subject to a variety of anthropogenic threats, but they often breed in remote colonies that are highly sensitive to disturbance. Aerial remote sensing can overcome some of these difficulties to census breeding sites and explore how local environmental factors influence important characteristics such as nest-site selection and chick survival. To this end, we used drone photography to map giant petrel nests, repeatedly evaluate chick survival and quantify-associated physical and biological characteristics of the landscape at two neighboring breeding sites on Humble Island and Elephant Rocks, along the western Antarctic Peninsula in January–March 2020. Nest sites occurred in areas with relatively high elevations, gentle slopes, and high wind exposure, and statistical models predicted suitable nest-site locations based on local spatial characteristics, explaining 72.8% of deviance at these sites. These findings demonstrate the efficacy of drones as a tool to identify, map, and monitor seabird nests, and to quantify important habitat associations that may constitute species preferences or sensitivities. These may, in turn, contextualize some of the diverse population trajectories observed for this species throughout the changing Antarctic environment. 

Palms play an outsized role in tropical forests and are important resources for humans and wildlife. A central question in tropical ecosystems is understanding palm distribution and abundance. However, accurately identifying and localizing palms in geospatial imagery presents significant challenges due to dense vegetation, overlapping canopies, and variable lighting conditions in mixed-forest landscapes. Addressing this, we introduce PalmProbNet, a probabilistic approach utilizing transfer learning to analyze high-resolution UAV-derived orthomosaic imagery, enabling the detection of palm trees within the dense canopy of the Ecuadorian Rainforest. This approach represents a substantial advancement in automated palm detection, effectively pinpointing palm presence and locality in mixed tropical rainforests. Our process begins by generating an orthomosaic image from UAV images, from which we extract and label palm and non-palm image patches in two distinct sizes. These patches are then used to train models with an identical architecture, consisting of an unaltered pre-trained ResNet-18 and a Multilayer Perceptron (MLP) with specifically trained parameters. Subsequently, PalmProbNet employs a sliding window technique on the landscape orthomosaic, using both small and large window sizes to generate a probability heatmap. This heatmap effectively visualizes the distribution of palms, showcasing the scalability and adaptability of our approach in various forest densities. Despite the challenging terrain, our method demonstrated remarkable performance, achieving an accuracy of 97.32% and a Cohen's κ of 94.59% in testing.