Format: ORAL

Barbara M. Neto-Bradley1, J. Antonio Guzmn Q.2, Jeannine Cavender-Bares2, David A. Coomes1

1 Department of Plant Sciences, University of Cambridge, UK 2 Department of Ecology, Evolution and Behavior, University of Minnesota, USA

Global biodiversity datasets contain key gaps, this is particularly obvious when considering taxa in the Global South. Natural History Collections have the potential to help fill these gaps, however many records are unidentified, precluding them from use. A need to non-destructively sample from these collections further limits their potential. We evaluate the added value gained by taking spectral reflectance measurements of leaf tissue from herbarium specimens. We use a previously undigitised set of 1539 herbarium specimens from the Herbarium at Kew to ask how well we can discriminate between species using spectral data. We also evaluate the feasibility of inferring traits from these herbarium derived spectra. Spectra can accurately be used to identify species, however further work is still needed to extend this application to trait estimates. We highlight that leaf reflectance spectra provide a novel and non-destructive way to add value to natural history collections and can be applied to fill systemic knowledge gaps.