Widespread sampling biases in herbaria revealed from large-scale digitization

1. Non-random collecting practices may bias conclusions drawn from analyses of herbarium records. Recent efforts to fully digitize and mobilize regional floras offer a timely opportunity to assess commonalities and differences in herbarium sampling biases. 2. We determined spatial, temporal, trait, phylogenetic, and collector biases in ~5 million herbarium records, representing three of the most complete digitized floras of the world: Australia (AU), South Africa (SA), and New England (NE). 3. We identified numerous shared and unique biases among these regions. Shared biases included specimens i) collected close to roads and herbaria; ii) collected more frequently during spring; iii) of threatened species collected less frequently; and iv) of close relatives collected in similar numbers. Regional differences included i) over-representation of graminoids in SA and AU and of annuals in AU; and ii) peak collection during the 1910s in NE, 1980s in SA, and 1990s in AU. Finally, in all regions, a disproportionately large percentage of specimens were collected by a few individuals. These mega-collectors, and their associated preferences and idiosyncrasies, may have shaped patterns of collection bias via 9founder effects9. 4. Studies using herbarium collections should account for sampling biases and future collecting efforts should avoid compounding these biases.