drmr: A Bayesian approach to Dynamic Range Models in R

This repository provides the R code and data to reproduce the results for the paper, "drmr: A Bayesian approach to Dynamic Range Models in R."

Installation

Before running any analysis, please follow these steps to set up the required environment.

1. Install CmdStan: This project depends on CmdStan (version >= 2.36). Please install it and the R package cmdstanr by following the official instructions.

2. Restore R Environment: This repository uses renv to manage R package dependencies (including the version of drmr in the pkg/ directory). To install all required packages, open this project in R and run:y

   renv::restore()

3. System Dependencies (Optional): If you wish to regenerate the raw environmental data for the red-bellied woodpecker case study, you must also have the bash interface to GDAL set up in your system. This is not required if you use the provided, pre-processed data.

Reproducing the Results

The analyses for the two case studies can be run using the scripts in the code/ directory.

Case Study 1: Summer Flounder

To reproduce the full analysis for the summer flounder, run the code/01-summer-flounder.R script.y

Case Study 2: Red-bellied Woodpecker

For convenience, the final processed dataset for this study is already provided at data/birds/processed.parquet.

• To run the analysis using the provided data, execute the code/04-rbw.R script.

• (Optional) To regenerate the processed data from raw sources, run the following scripts in order:

  1. code/02-download-bbs.R: Downloads Breeding Bird Survey (BBS) data via the bbsBayes2 package.
  2. chelsa-download.sh: A bash script to download and crop CHELSA air temperature data.
  3. code/03-chelsa-env4bbs.R: Aggregates the temperature data and creates the final processed.parquet file.

Notes on Reproducibility

Please be aware that results from CmdStan may vary slightly across different operating systems and hardware platforms due to minor differences in random number generator implementations. Full reproducibility is only guaranteed when using an identical computational environment.

System information

The analyses were conducted on a macOS machine with the following specifications:

• Processor: Apple M2 Pro
• Memory: 16GB RAM

The R session information is provided below for full reproducibility.

> sessionInfo()
#> R version 4.4.1 (2024-06-14)
#> Platform: aarch64-apple-darwin20
#> Running under: macOS 15.5
#>
#> Matrix products: default
#> BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Vers#> ions/A/libBLAS.dylib 
#> LAPACK: /Library/Frameworks/R.framework/Versions/4.4-arm64/Resources/lib/libRlapack.dylib;  LAPACK version 3.12.0