Too few, too many, or just right? Optimizing sample sizes for population-level inferences in animal tracking projects

Table of contents

1. Introduction
2. Tutorials
3. Directory structure
4. Scripts description
5. Citations
6. License

Introduction

This repository is a companion piece to the manuscript “Too few, too many, or just right? Optimizing sample sizes for population-level inferences in animal tracking projects”, currently available as a preprint here.

The main workflow presented in the manuscript provides a comprehensive approach for optimizing sample sizes in animal tracking studies, balancing sampling duration, sampling interval, and the number of sampled individuals to ensure robust, unbiased population-level inferences for home range, and speed & distance estimation. By integrating robust, sampling-insensitive analytical methods and accounting for uncertainty and logistical constraints (including fix success, location error, device malfunctions, and individual variation), this workflow guides researchers in designing effective studies or evaluating existing data.

The workflow is implemented in the movedesign Shiny application and R package, available on CRAN and GitHub.

Tutorials

To install the stable version of movedesign from CRAN:

install.packages("movedesign")

To install the most recent development version directly from GitHub:

install.packages("remotes")
remotes::install_github("ecoisilva/movedesign")

Run Shiny application:

To launch movedesign, load the library and run the following command in your R console:

library(movedesign)
movedesign::run_app()

A step-by-step tutorial for the Shiny interface is available here.

Run in R console:

The full workflow can also be run directly in the R console. Users start by fitting continuous-time movement models to empirical datasets, then pass those models into md_prepare() to configure candidate designs defined by three parameters: number of tagged individuals, sampling duration, and sampling interval. Two contrasting designs are compared: one short and fine-resolution, one long and coarse. By running simulations with md_run(), previewing estimation error with md_plot_preview() and md_compare_preview(), then scaling up to multiple replicates with md_replicate() for stable inference. Finally, users can compare and rank designs using md_compare().

As an example, you can follow along the tutorial available here.

Directory structure

The directory structure below provides all code required to reproduce the data and figures from the manuscript. For guidance on applying the workflow to your own data, refer to the tutorials in the Tutorials section instead.

studydesign_ms/
│-- cluster/              # HPC bash files and scripts
│-- data/                 # Data files
│-- documentation/        # Tutorial and supplementary materials
│-- figures/              # Generated figures
│-- outputs/              # Processed outputs
│-- R/                    # R scripts
│-- renv/                 # `renv` package management directory
│-- renv.lock             # Lock file for dependencies
│-- global.R              # Global settings and objects
│-- studydesign_ms.Rproj  # RStudio project file

Scripts description

The 📁 cluster folder stores scripts for HPC jobs:

• cluster/create_job_sh.R: generates job submission scripts.
• cluster/mean_hr.R: runs simulations for mean home range.
• cluster/mean_speed.R: runs simulations for mean movement speed.
• cluster/meta_resampled.R: runs meta-analyses and resampling approach.

The 📁 R folder contains the other main scripts:

• R/plotting_case-study.R: generates output plots for a case study.
• R/running_case-study_loocv.R: runs LOOCV analyses.
• R/plotting_case-study_loocv.R: generates LOOCV output plots for a case study.
• R/plotting_case-study_resampled.R: generates resample plots for a case study.
• R/plotting_sims_resampled.R: generates resample plots for the simulations.

The 📁 R/functions subfolder contains helper functions used throughout the project.

Citations

If you use this workflow or the movedesign package in your work, please cite:

Silva, I., Fleming, C. H., Noonan, M. J., Fagan, W. F., & Calabrese, J. M. (2025). Too few, too many, or just right? Optimizing sample sizes for population-level inferences in animal tracking projects. BioRxiv. DOI: https://doi.org/10.1101/2025.07.30.667390

Silva, I., Fleming, C. H., Noonan, M. J., Fagan, W. F., & Calabrese, J. M. (2023). movedesign: Shiny R app to evaluate sampling design for animal movement studies. Methods in Ecology and Evolution, 14(9), 2216–2225. DOI: https://doi.org/10.1111/2041-210X.14153

License