Year of Award
2026
Document Type
Thesis
Degree Type
Master of Science (MS)
Degree Name
Wildlife Biology
Department or School/College
W.A. Franke College of Forestry and Conservation
Committee Chair
Dr. Joshua J. Millspaugh
Commitee Members
Dr. Daniel P. Walsh, Dr. Thomas V. Riecke
Keywords
Bird Radar, Bird Strikes, Human Wildlife Conflict
Subject Categories
Ornithology
Abstract
Monitoring the migration and movement of bird populations is important for avian conservation. Increasingly, radar systems are being used to understand patterns in movement activity and assess risk of bird strikes with human infrastructure, such as airplanes and wind turbines. While radar systems excel at monitoring bird activity, current systems cannot identify avian targets to finer levels of classification, such as morphological groups or size class, limiting their utility for avian monitoring. Additionally, there are questions surrounding how avian activity is influenced by weather and timing to better inform bird strike risk management at airfields. We worked with radar systems at two US Air Force bases, Ellsworth Air Force Base, SD (EAFB) and Offutt Air Force Base, NE (OAFB) to address these limitations and questions. In Chapter 1, we built machine learning classification models to identify tracks to different morphological and biomass groups of birds using a dataset of tracks identified to species and quantity at each base. We were able to successfully identify unclassified radar tracks to bird morphological groups including songbirds, waterfowl, raptors, gull (at OAFB only) and herons (at EAFB only) using neural networks. We were also able to classify tracks to four levels of track biomass at both bases. Models were base- and equipment-specific, indicating future modeling efforts will require further collection of identified track datasets at a new radar location. In Chapter 2, we built models explaining avian activity as a function of weather and temporal covariates across multiple seasons to understand what factors most influence the intensity of avian activity on an hourly scale, and multiple monitoring scales to see if the effects of covariates on avian activity differed by monitoring scale. Across bases, hour of day, and wind speed and direction played important roles in influencing bird activity, though the strength and direction of effects changed seasonally. We did not find a difference in these effects by monitoring scale at EAFB but did see a difference at OAFB, especially in summer and winter. Our results highlight the importance of building base specific models explaining bird strike risk to aircraft.
Recommended Citation
Salvo, Marco, "IMPROVING DEDICATED BIRD RADAR SYSTEMS THROUGH AVIAN TARGET CLASSIFICATION AND BIRD STRIKE RISK ASSESSMENT" (2026). Graduate Student Theses, Dissertations, & Professional Papers. 12603.
https://scholarworks.umt.edu/etd/12603
© Copyright 2026 Marco Salvo