geoorigins: A new method and r package for trait mapping and geographic provenancing of specimens without categorical constraints

Biologists often seek to geographically provenance organisms using their traits. This is typically achieved by defining spatial groups using distinct patterns of trait variation. Here, we present a new spatial provenancing and trait boundary identification methodology, based on correlations between geographic and trait distances that require no a priori group assumptions. We apply this to three datasets where spatial provenance is sought: morphological rat and vole dentition data (human commensal translocation datasets); and birdsong data (cultural transmission dataset). We also present the results of cross‐validation testing. Spatial provenancing is possible with differing degrees of accuracy for each dataset, with birdsong providing the most accurate geographic origin (identifying an average spatial region of 0.22 km2 as the area of origin with 99.9% confidence). Our method has a wide range of potential applications to diverse data types—including phenotypic, genetic and cultural—to identify trait boundaries and spatially provenance the origin of unknown or translocated specimens where trait differences are geographically structured and correlated with spatial separation.