In an effort to monitor biodiversity trends, greater efforts are being made worldwide to assess biodiversity patterns over large scales. To do this, scientists rely on species distribution models (SDMs), which make predictions of species' geographical ranges based on species data and environmental variables. With these models, scientists can make predictions of habitat suitability under different global change scenarios and tailor management and conservation efforts accordingly.

The international Group on Earth Observations Biodiversity Observation Network (GEO BON), which pools resources and researchers from across the globe, has recently conceptualized "essential biodiversity variables" to standardize the collection and coordination of biodiversity data, and many of these variables can be made with SDMs. Most cutting-edge SDM techniques are implemented in R, a popular statistical programming language, and many new tools have surfaced in recent years in the form of R packages. But researchers often get overwhelmed by the plethora of R packages out there, wondering, 'Which one should I use for my research?'

In a new paper, Jamie M. Kass, associate professor and head of the Macroecology Lab at Tohoku University's Graduate School of Life Sciences, argues that SDM workflows benefit highly from use of multiple packages. Kass, who has helped develop several R packages for SDMs - including ENMeval (which fine-tunes machine-learning SDMs) and wallace (a user-friendly application for SDM workflows) - teamed up with experts worldwide to create a guide for using multiple R package tools effectively and in innovative ways.

The team introduced a new R meta-package called sdmverse, which catalogs R packages for SDMs by the functions they offer and provides visualization features to help researchers understand how they relate to each other. They also contributed three real-world case studies in R showing how combining tools can broaden the diversity of analyses possible and help meet more methodological standards for the field.

"New tools help science move forward, but they can also be overwhelming," says Kass. "We wanted to create a roadmap that shows researchers how to navigate these tools and use them together for better biodiversity modeling."

By following their approach, researchers can improve accuracy, tackle a wider range of questions, and contribute to stronger biodiversity assessments worldwide. As environmental challenges grow, using the best available tools--together - will be essential for tracking trends in biodiversity and protecting nature.