Global genetic diversity maps

High-resolution global maps of microbial genetic diversity derived from over 25 billion environmental DNA (eDNA) reads. Methodology: More than 250,000 eDNA sequencing runs from soil samples were curated from over 1000 large-scale public projects worldwide. Nucleotide diversity (π) between denoised amplicon sequence variants (ASVs) of 16S rRNA and ITS markers was computed across bacterial and fungal OTUs at the phylum level, respectively. A geospatial modeling pipeline developed by the Crowther Lab (van den Hoogen et al., 2021) was used to train a random forest model of nucleotide diversity using a k-fold cross-validation approach and 20 climatic, environmental, and soil physicochemical covariates as predictive features. The resulting maps demonstrate the feasibility and utility of integrating eDNA data with geospatial earth observations to establish a robust, scalable, and standardized framework for monitoring global genetic diversity and addressing longstanding limitations of conventional biodiversity assessments. Units: Nucleotide diversity represents the average proportion of differences between any two randomly selected DNA sequences, and the values range from 0 to 1. Spatial resolution: 30 arc seconds (~1km at the equator; exact resolution varies globally due to map projection, with pixel size decreasing towards the poles) Geographic coverage: Global (except Antarctica); terrestrial Temporal coverage: 2010-2025 Updates: This is the first version of the genetic diversity maps demonstrating the proof-of-concept. Major updates to the maps are expected with increased eDNA sampling coverage, spatial resolution, temporal resolution, and taxonomic diversity.