A Lattice and Random Intermediate Point Sampling Design for Animal Movement

TL;DR

This paper introduces the LARI sampling design, combining regular and random sampling intervals, which improves parameter estimation and data prediction in animal movement studies compared to traditional regular sampling.

Contribution

The paper proposes a novel LARI sampling design that integrates regular and random sampling intervals to enhance data efficiency in animal movement research.

Findings

LARI sampling improves parameter estimation accuracy.

LARI sampling enhances prediction of missing data.

Regular sampling yields higher precision when parameters are well-estimated.

Abstract

Animal movement studies have become ubiquitous in animal ecology for estimation of space use and analysis of movement behavior. In these studies, animal movement data are primarily collected at regular time intervals. We propose an irregular sampling design which could lead to greater efficiency and information gain in animal movement studies. Our novel sampling design, called lattice and random intermediate point (LARI), combines samples at regular and random time intervals. We compare the LARI sampling design to regular sampling designs in an example with common black carpenter ant location data, an example with guppy location data, and a simulation study of movement with a point of attraction. We modify a general stochastic differential equation model to allow for irregular time intervals and use this framework to compare sampling designs. When parameters are estimated reasonably well, regular sampling results in greater precision and accuracy in prediction of missing data. However, in each of the data and simulation examples explored in this paper, LARI sampling results in more accurate and precise parameter estimation, and thus better prediction of missing data as well. This result suggests that researchers might gain greater insight into underlying animal movement processes by choosing LARI sampling over regular sampling.