Master's Thesis
Citizen vs. Professional Data for Seabird Monitoring: A Machine Learning Perspective
— 2025
Key information
Authors:
Supervisors:
Published in
May 19, 2025
Abstract
This thesis explores the integration of citizen science and institutional professional data to enhance seabird monitoring along the Portuguese coast and to reduce the dependency on resource-intensive institutional surveys. Using extensive eBird observations and structured data from the RAM Days program coordinated by SPEA, a machine learning framework based on Random Forests was developed to model species detectability and abundance across coastal observation sites. Both datasets were extensively preprocessed, spatially harmonized and enriched with environmental predictors. Challenges such as class imbalance and observer bias were addressed through targeted sampling strategies and post-calibration techniques. Models were trained and evaluated for nine marine bird species, with institutional data serving as a benchmark to validate and correct citizen science predictions. Results showed that eBird-based models performed well in species classification and, after appropriate calibration, produced abundance estimates that aligned with institutional data in specific contexts. In particular, certain species-locality combinations exhibited strong agreement, demonstrating the potential for reliable integration. These findings suggest that, when adequately corrected, citizen science data can function as a valuable complementary tool, especially in areas or seasons where institutional coverage is limited. The study presents a scalable, transferable framework for integrating structured and opportunistic biodiversity data, emphasizing its utility for conservation purposes under resource constraints. By merging ecological knowledge with machine learning, the approach supports more comprehensive, high-resolution monitoring and provides a practical pathway for data-informed conservation planning in coastal and potentially other ecological systems.
Publication details
Authors in the community:
Ana Rita Dos Reis Duarte
ist195531
Supervisors of this institution:
Fields of Science and Technology (FOS)
electrical-engineering-electronic-engineering-information-engineering - Electrical engineering, electronic engineering, information engineering
Publication language (ISO code)
eng - English
Rights type:
Embargo lifted
Date available:
March 29, 2026
Institution name
Instituto Superior Técnico