Bayesian approach for predicting photogrammetric uncertainty in morphometric measurements derived from drones

Increasingly, drone-based photogrammetry has been used to measure size and body condition changes in marine megafauna. A broad range of platforms, sensors, altimeters are being applied for these purposes, but there is no unified way predict photogrammetric uncertainty across this methodological spectrum. As such, it difficult make robust comparisons studies, disrupting collaborations amongst researchers using platforms with varying levels measurement accuracy. Here we built off previous studies quantifying an experimental approach train a Bayesian statistical model known-sized object floating at the water’s surface quantify how error scales altitude several different drones equipped cameras, focal length lenses, altimeters. We then fitted distributions estimate age classes unknown-sized humpback whales Megaptera novaeangliae , as well population-level morphological relationship between rostrum blowhole distance total Antarctic minke Balaenoptera bonaerensis . This framework jointly estimates errors from measurements multiple observations accounts altitudes measured both barometers laser while incorporating specific each. outputs posterior predictive distribution around allows construction highest density intervals define uncertainty, which one probabilistic statements stronger inferences pertaining morphometric features critical understanding life history patterns potential impacts anthropogenically altered habitats.