Abstract
BackgroundIn recent years the field of movement ecology has been revolutionized by our ability to collect high-accuracy, fine scale telemetry data from individual animals and groups. This growth in our data collection capacity has led to the development of statistical techniques that integrate telemetry data with random walk models to infer key parameters of the movement dynamics. While much progress has been made in the use of these models, several challenges remain. Notably robust and scalable methods are required for quantifying parameter uncertainty, coping with intermittent location fixes, and analysing the very large volumes of data being generated.MethodsIn this work we implement a novel approach to movement modelling through the use of multilevel Gaussian processes. The hierarchical structure of the method enables the inference of continuous latent behavioural states underlying movement processes. For efficient inference on large data sets, we approximate the full likelihood using trajectory segmentation and sample from posterior distributions using gradient-based Markov chain Monte Carlo methods.ResultsWhile formally equivalent to many continuous-time movement models, our Gaussian process approach provides flexible, powerful models that can detect multiscale patterns and trends in movement trajectory data. We illustrate a further advantage to our approach in that inference can be performed using highly efficient, GPU-accelerated machine learning libraries.ConclusionsMultilevel Gaussian process models offer efficient inference for large-volume movement data sets, along with the fitting of complex flexible models. Applications of this approach include inferring the mean location of a migration route and quantifying significant changes, detecting diurnal activity patterns, or identifying the onset of directed persistent movements.
Highlights
Animal movement is a fundamental ecological process that influences the dynamics of ecosystems across multiple spatiotemporal scales
Our regression model follows the 2-dimensional version of Eqs. 1 and 2, where we model each output of the multivariate Gaussian process as two univariate GPs
We firstly detect daily activity patterns in a simulated individual organism, we infer changes to a seasonal migration route where an annual movement pattern is shifting over a longer timescale
Summary
Animal movement is a fundamental ecological process that influences the dynamics of ecosystems across multiple spatiotemporal scales. Over recent years there has been a rapid advance in our ability to collect data on the movement behaviour of many animal species [4, 5] and this has led to the development of sophisticated statistical methods to analyse these data [6]. In recent years the field of movement ecology has been revolutionized by our ability to collect high-accuracy, fine scale telemetry data from individual animals and groups. This growth in our data collection capacity has led to the development of statistical techniques that integrate telemetry data with random walk models to infer key parameters of the movement dynamics. Robust and scalable methods are required for quantifying parameter uncertainty, coping with intermittent location fixes, and analysing the very large volumes of data being generated
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have