Widespread sampling biases in herbaria revealed from large‐scale digitization

• DOI: 10.1111/nph.14855
• Publication Date: 2017-10-30T10:00:35Z
• Authors: Barnabas H. Daru; Daniel S. Park; Richard B. Primack; Charles G. Willis; David S. Barrington; Timothy J. S. Whi...; Tristram G. Seidler; Patrick W. Sweeney; David R. Foster; Aaron M. Ellison; Charles C. Davis
• Keywords: Herbarium; Digitization; Sampling bias

Summary Nonrandom collecting practices may bias conclusions drawn from analyses of herbarium records. Recent efforts to fully digitize and mobilize regional floras online offer a timely opportunity assess commonalities differences in sampling biases. We determined spatial, temporal, trait, phylogenetic, collector biases c . 5 million records, representing three the most complete digitized world: Australia ( AU ), South Africa SA New England, USA NE ). identified numerous shared unique among these regions. Shared included specimens collected close roads herbaria; more frequently during biological spring summer; threatened species less frequently; relatives similar numbers. Regional overrepresentation graminoids annuals ; peak collection 1910s , 1980s 1990s Finally, all regions, disproportionately large percentage were by very few individuals. hypothesize that mega‐collectors, with their associated preferences idiosyncrasies, shaped patterns via ‘founder effects’. Studies using collections should account for biases, future avoid compounding extent possible.