momentuHMM: <scp>R</scp> package for generalized hidden Markov models of animal movement

Abstract Discrete‐time hidden Markov models (HMMs) have become an immensely popular tool for inferring latent animal behaviours from telemetry data. While movement HMMs typically rely solely on location data (e.g. step length and turning angle), auxiliary biotelemetry and environmental data are powerful and readily‐available resources for incorporating much more ecological and behavioural realism. However, complex movement or observation process models often necessitate custom and computationally demanding HMM model‐fitting techniques that are impractical for most practitioners, and there is a paucity of generalized user‐friendly software available for implementing multivariate HMMs of animal movement. Here, we introduce an open‐source R package, momentuHMM , that addresses many of the deficiencies in existing HMM software. Features include: (1) data pre‐processing and visualization; (2) user‐specified probability distributions for an unlimited number of data streams and latent behaviour states; (3) biased and correlated random walk movement models, including dynamic “activity centres” associated with attractive or repulsive forces; (4) user‐specified design matrices and constraints for covariate modelling of parameters using formulas familiar to most R users; (5) multiple imputation methods that account for measurement error and temporally irregular or missing data; (6) seamless integration of spatio‐temporal covariate raster data; (7) cosinor and spline models for cyclical and other complicated patterns; (8) model checking and selection; and (9) simulation. After providing an overview of the main features of the package, we demonstrate some of the capabilities of momentuHMM using real‐world examples. These include models for cyclical movement patterns of African elephants, foraging trips of northern fur seals, loggerhead turtle movements relative to ocean surface currents, and grey seal movements among three activity centres. momentuHMM considerably extends the capabilities of existing HMM software while accounting for common challenges associated with telemetry data. It therefore facilitates more realistic hypothesis‐driven animal movement analyses that have hitherto been largely inaccessible to non‐statisticians. While motivated by telemetry data, the package can be used for analysing any type of data that is amenable to HMMs. Practitioners interested in additional features are encouraged to contact the authors.