A Semi-Automated Method for Estimating Adélie Penguin Colony Abundance from a Fusion of Multispectral and Thermal Imagery Collected with Unoccupied Aircraft Systems

Abstract

Monitoring Adélie penguin (Pygoscelis adeliae) populations on the Western Antarctic Peninsula (WAP) provides information about the health of the species and the WAP marine ecosystem itself. In January 2017, surveys of Adélie penguin colonies at Avian Island and Torgersen Island off the WAP were conducted via unoccupied aircraft systems (UAS) collecting optical Red Green Blue (RGB), thermal, and multispectral imagery. A semi-automated workflow to count individual penguins using a fusion of multispectral and thermal imagery was developed and combined into an ArcGIS workflow. This workflow isolates colonies using multispectral imagery and detects and counts individuals by thermal signatures. Two analysts conducted manual counts from synoptic RGB UAS imagery. The automated system deviated from analyst counts by −3.96% on Avian Island and by 17.83% on Torgersen Island. However, colony-by-colony comparisons revealed that the greatest deviations occurred at larger colonies. Matched pairs analysis revealed no significant differences between automated and manual counts at both locations (p > 0.31) and linear regressions of colony sizes from both methods revealed significant positive relationships approaching unity (p < 0.0002. R2 = 0.91). These results indicate that combining UAS surveys with sensor fusion techniques and semi-automated workflows provide efficient and accurate methods for monitoring seabird colonies in remote environments.

Applications of unmanned aerial vehicles in Antarctic environmental research

“…However, the distribution and population dynamics of species such as the Chinstrap penguin are not well understood, mainly because they nest in remote and rugged areas, on-the-ground census work is difficult and sporadic 25 . As demonstrated for Adelia penguins 26 the use of thermal imagery would allow reliable population estimates of Chinstrap penguins. Even, the use of RGB aerial images for animal counting would be far more accurate than from land-based surveys.…”

Automating sandhill crane counts from nocturnal thermal aerial imagery using deep learning

“…Methodologies for application of thermal imagery for large mammal monitoring are relatively well established, but application of this technology to bird monitoring over large areas has been considered problematic because of the coarse resolution of thermal sensors and the small body size of birds (Chabot & Francis, 2016). More recent studies with state‐of‐the‐art sensors have utilized thermal imagery to count penguins (Bird et al., 2020) and identify individual nesting birds (McKellar et al., 2021; Santangeli et al., 2020) over relatively limited areas ( i.e . tens to hundreds of meters).…”

“…Commonly, a spectral threshold approach is taken to distinguish and count individual animals in imagery (Bird et al., 2020; Chabot & Francis, 2016). However, this technique requires substantial manual calibration to account for diverse thermal signatures and spectral variation, potentially limiting applicability of the resulting algorithm.…”

UAVs for Science in Antarctica

“…The counting of tens of thousands of penguin adults in 14 colonies located along the northern coasts of the adjacent Nelson and King George islands, together with data derived from terrestrial time-lapse imagery, are the basis for Pfeifer et al ( 2019) [129] to model the number of occupied nests and verify their decline since the 1980s. The fusion of multispectral and thermal imagery acquired after UAV surveys in early 2017 by Bird et al (2020) [130] were used to improve the detection of penguins in Avian and Torgersen Islands in the WAP in semi-automated workflows. The survey of two large colonies in Ross Island was achieved by Shah et al (2020) [131] in the 2019-2020 summer with four multi-rotor UAVs operating simultaneously.…”