@misc{10498/40520,
year = {2026},
url = {http://hdl.handle.net/10498/40520},
abstract = {Recent advances in machine learning have accelerated automated species detection across diverse ecological domains, enabling large-scale, non-invasive monitoring of biodiversity. In ornithological research, the combination of passive acoustic monitoring (PAM) and rapidly-developing novel identification tools such as BirdNET-a deep learning-based sound recognition algorithm-offers new opportunities for surveying vocally active bird communities. Here, we present the first worldwide evaluation of BirdNET using 4224 one-minute recordings from 67 sites across all continents annotated by local experts. More specifically, we assessed the capacity of BirdNET to accurately identify individual vocalizations and characterize bird communities based on the automated analysis of passively collected soundscapes. We further analyzed how its performance varies across continents, biomes, species, and minimum confidence thresholds. The proportion of correct BirdNET predictions (precision) was generally high and consistent across continents (range: 0.57-0.71) and biomes (range: 0.55-0.76). In contrast, the proportion of vocalizations successfully detected (recall) was generally lower and more heterogeneous across continents (range: 0.24-0.52) and biomes (range: 0.34-0.72), reflecting differences in species coverage and local ecological context. BirdNET predictive power, as measured by the Precision-Recall Area Under the Curve (PR AUC; higher values indicating better performance), was highest in North America, Oceania, and Europe (range: 0.16-0.23), moderate in Central/South America (0.13), and lowest in Africa and Asia (range: 0.03-0.04). Species-specific analyses revealed substantial heterogeneity in detection accuracy, with optimal confidence thresholds varying widely by species and analytical goal. Our results establish a global reference point for BirdNET reliability and highlight where algorithmic refinement and expanded acoustic sampling are most needed.},
publisher = {ELSEVIER},
keywords = {Confidence threshold},
keywords = {Automated detection},
keywords = {Passive acoustic monitoring},
keywords = {Bird communities},
keywords = {BirdNET},
keywords = {Deep learning},
title = {A global assessment of BirdNET performance: Differences among continents, biomes, and species},
doi = {10.1016/J.ECOLIND.2025.114550},
author = {Funosas, David and Sebastián-González, E. and Morant, Jon and Marín Gómez, O.H. and Mendoza, Irene and Mohedano Muñoz, Miguel A. and Santamaría, E. and Bastianelli, Giulia and Marquez Rodríguez, Alba and Budka, Michał and Bota, G. and Alonso-Moya, C.D. and De la Peña Rubio, José M. and de la Morena, E.L.G. and Santa-Cruz, M. and de la Nava, Pablo and Fernández Tizón, Mario and Sánchez Mateos, Hugo and Barrero, Adrián and Traba, Juan and Osiejuk, Tomasz S. and Hart, Patrick J and Navine, Amanda K. and Montoya Muñoz, Andrés F. and de Araujo, Carlos B. and Rosa, G.L.M. and Torres, I.M.D. and Catalano, Ana L. and Simões, C.R. and Llusia, Diego and Morales, M.B. and Acebes, Pablo and Medina, J.A. and Brown, N. and Astaras, Christos and Karmiris, I. and Navarrete, E. and Cauchoix, Maxime and Barbaro, Luc and Arend, D. and Müeller, Sandra and González-García, F. and González Romero, Alberto and Mammides, Christos and Pontikis, Michaelangelo and Jacuzzi, Giordano and Olden, Julian D. and Bombaci, Sara P. and Marcacci, Gabriel and Jacot, Alain and Zurano, J.P. and Gangenova, Elena and Varela, D. and di Sallo, Facundo and Zurita, Gustavo A. and Atemasov, Andrey and Tremblay, Junior A. and Lamarre, V. and Hutschenreiter, A. and Monroy Ojeda, Alan and Díaz Vallejo, Mauricio and Chaparro Herrera, Sergio and Briers, Robert A. and Sousa Lima, Renata and Pinheiro, Thiago and Da Silva, W.C. and Calvente, Alice and Paz, R.V. and Salustio Gomes, Carlos and Oliveira Junior, Dorgival D. and Lima Santos, Cicero S. and Pichorim, Mauro and Molin, A.D. and Antonelli, Alexandre and Gogoleva, Svetlana and Palko, Igo and Trong, Hiếu V. and Duarte, M.H.L. and dos Santos Saturnino, Natalia and Silva, S.R. and Rainho, Ana and Lopes, P. and Schuchmann, Karl L. and Marques, M.I. and de Oliverira Tissiani, A.S. and Littlewood, Nick A. and Tuanmu, Mao-Ning and Kepfer Rojas, Sebastian and Aguilera, A.L. and Brotons, Lluís and Feldman, M.J. and Imbeau, Louis and Panwar, Pooja and Weed, Aaron S. and Dehwal, Anant and Attisano, Alfredo and Theuerkauf, Jörn and Goodale, Eben and Darras, K.F.A. and Pérez Granados, Cristian},
}