The Usability of a Small Museum for Research, a Look Inside the Witte Museum Herbarium

Herbaria, a collection of preserved plant specimens with associated information are important because they offer valuable reference materials for both education and research. However, often they are underutilized because the information they contain is not readily available to researchers particularly in the case of those housed in small, regional museums. Such herbaria are likely to harbor data collected over time frames and geographic regions underrepresented in existing publicly available datasets, and thus may provide critical information for botanical research, including that detailing the historic distribution of plants. However, the value of herbaria from small, regional museums for botanical research in this capacity has been largely ignored. Understanding the value of herbaria in this capacity also includes assessment of the usability of specimens (e.g., how many specimens from a herbarium contain complete and relevant data to permit a scientific study) and the potential biases associated with the specimens that can influence conclusions drawn from such studies. The overall objective of this study was to assess the usability of a herbarium from the Witte Museum, a small regional museum in San Antonio, Texas for botanical research. Specifically, my objectives were to (1) determine collection statistics for the herbarium, (2) identify collection biases in the herbarium, and (3) determine if the herbarium increases knowledge of the historic distribution of common plant families. To address my objectives, first I digitized specimens by transcribing information from specimen labels from the herbarium and creating a herbarium dataset. I evaluated the completeness of data provided on specimen labels after correcting for changes in scientific nomenclature and determining geographic coordinates based on locality descriptions. The resulting dataset contained taxonomic information (order, family, genus, species), location (locality descriptions and geographic coordinates), collector, and collection date. As a second step, I evaluated geographic, temporal, and taxonomic biases in the dataset. My results suggest that of 4,807 specimens in the Witte Herbarium, the majority were collected in Bexar County, Texas (26.17% of the herbarium) between 1920-1929 (38.02% of the herbarium) and that specimens from the Asteraceae family were collected more frequently than specimens from other families (28.40% of the herbarium), suggesting temporal, geographic, and taxonomic biases in the herbarium. Finally, I evaluated the efficacy of the herbarium to increase knowledge about the distribution of the five most represented plant families (Asteraceae, Fabaceae, Poaceae, Brassicaceae, Laminaceae) using specimens collected between 1920-1929 (the peak period of specimen collection). Specifically, I compared Areas of Occupancy (AOOs) described by herbarium data with AOOs described using publicly available point occurrence data from the Global Biodiversity Information Facility (GBIF). I found that the AOOs calculated using the Witte Herbarium were larger than those described by the GBIF data for four of the five families, suggesting that the herbarium increased knowledge of the distribution of these families between 1920-1929 when considering publicly available data. Although my results suggest that biases can occur in small herbaria, they also suggest that small herbaria harbor valuable information that can expand existing knowledge, and thus that small herbaria are important contributors to ecological data.