Research overview: fix-success bias under canopy and rugged terrain

GPS collars do not always acquire a position, and crucially, the probability of failure is not random: dense canopy, steep terrain, and an animal resting in a cave or thicket all suppress fixes. This overview explains why such non-random missingness biases analyses of habitat use, since the very habitats cats prefer for cover may be underrepresented in the data. Researchers quantify the problem with stationary test collars placed across habitat types and correct it by modeling fix success as a function of environmental covariates. For forest cats like tigers, clouded leopards, and jaguars, addressing this bias is essential. The topic links engineering reality to ecological inference and is a recurring caution in carnivore telemetry.