Theme 4: Wildlife--Oral Sessions

Description

Grassland bird populations can be good indicator species of ecosystem health. However, their populations are declining at greater rates than any other group of birds. A well-established method of monitoring rapidly disappearing bird populations is by locating and identifying active nests. Studies quantifying grassland birds tend to have low statistical power due to low sample sizes, high labor costs, and high levels of disturbance - associated with difficulty finding nests. However, advances in small unmanned aerial systems (sUAS) and thermographic imaging technologies have the potential to improve efficiency and accuracy of locating nests, while causing minimal disruption. Early research has evaluated nest detectability using a thermal imaging system equipped to a sUAS. The sUAS was flown at three different altitudes to detect simulated nests at incremental depths in monoculture grass stand canopies. This study evaluated nest detection accuracy using visual assessment of two different types of thermal imagery. The first type of imagery used third-party software to create a stitched thermal map of the research area, while the second method utilized real-time video feed from the thermal sensor to identify simulated nest locations. Both methodologies were tested in a blind evaluation, using five evaluators and two replications. Results from this study have suggested that mapping software does not optimize nest detectability and identification, and the analysis of videos proves to be a much more precise way to detect and identify nests.

Share

COinS
 

Assessment of Thermographic Imaging Systems on Small Unmanned Aerial Systems (sUAS) to Identify Artificial Grassland Bird Nests

Grassland bird populations can be good indicator species of ecosystem health. However, their populations are declining at greater rates than any other group of birds. A well-established method of monitoring rapidly disappearing bird populations is by locating and identifying active nests. Studies quantifying grassland birds tend to have low statistical power due to low sample sizes, high labor costs, and high levels of disturbance - associated with difficulty finding nests. However, advances in small unmanned aerial systems (sUAS) and thermographic imaging technologies have the potential to improve efficiency and accuracy of locating nests, while causing minimal disruption. Early research has evaluated nest detectability using a thermal imaging system equipped to a sUAS. The sUAS was flown at three different altitudes to detect simulated nests at incremental depths in monoculture grass stand canopies. This study evaluated nest detection accuracy using visual assessment of two different types of thermal imagery. The first type of imagery used third-party software to create a stitched thermal map of the research area, while the second method utilized real-time video feed from the thermal sensor to identify simulated nest locations. Both methodologies were tested in a blind evaluation, using five evaluators and two replications. Results from this study have suggested that mapping software does not optimize nest detectability and identification, and the analysis of videos proves to be a much more precise way to detect and identify nests.