Elevational gradient analyses and the use of historical museum specimens: a cautionary tale

摘要: Aim  The value of biodiversity informatics rests upon the capacity to assess data quality. Yet as these methods have developed, investigating quality underlying specimen has largely been neglected. Using an exceptionally large, densely sampled set for non-flying small mammals Utah, I evaluate measures uncertainty associated with georeferenced localities and illustrate implications uncritical incorporation in analysis patterns species richness range overlap along elevational gradients. Location  USA, emphasis on Uinta Mountains. Methods  Employing from Mammal Networked Information System (MaNIS), converted estimates areal into using a geographic information system (GIS). Examining both measures, develop criteria including analyses gradients. Mountains test case, then examine gradient, without accounting quality. Results  GIS, provide framework post-hoc 3-dimensional georeferencing demonstrate collector-recorded elevations valuable technique detecting potential errors georeferencing. established evaluating when analysing case reduced number by 44% specimens 22%. Decreasing sample size this manner resulted subsequent removal one analysis. With quality, pattern gradient was hump-shaped peak at about mid-elevation, between 2300 2600 m. In contrast, frequencies different pair-wise among differed significantly not accounted for. Main conclusions  These results indicate that failing spatial error did alter shape observed nor species’ first last occurrences. However, it yield misleading community composition within given interval, well species. Patterns are often used infer processes distributions, suggesting failure account may interpretations process perceived distribution. is necessary component incorporating informatics.