Disentangling the role of remotely sensed spectral heterogeneity as a proxy for North American plant species richness

• Published in: Community ecology Vol. 15; no. 1; pp. 37 - 43
• Main Authors: Rocchini, D, L. Dadalt, L. Delucchi, M. Neteler, M.W. Palmer
• Format: Journal Article
• Language: English
• Published: Cham Akadémiai Kiadà 01-06-2014; AKADÉMIAI KIADÓ; Springer International Publishing
• Subjects: Animal Ecology; Area effects; Biodiversity; Community & Population Ecology; data collection; ecosystems; Flora; Hexagons; Life Sciences; Marine & Freshwater Sciences; Microbial Ecology; Plant Ecology; Remote sensing; spatial data; Species diversity; Spectral reflectance; Spectral resolution; Spectral variation hypothesis; Statistical variance; Synecology; variance; Variance partitioning; Vascular plants; North America; Species diversity; Spectral variation hypothesis; Variance partitioning; Area effects
• ISSN: 1585-8553; 1588-2756
• DOI: 10.1556/ComEc.15.2014.1.4

Due to the difficulties of field-based species data collection at wide spatial scales, remotely sensed spectral diversity has been advocated as one of the most effective proxies of ecosystem and species diversity. It is widely accepted that the relationship between species and spectral diversity is scale dependent. However, few studies have evaluated the impacts of scale on species diversity estimates from remote sensing data. In this paper we tested the species versus spectral relationship over very large scales (extents) with a varying spatial grain using floristic data of North America. Spectral diversity explained a low amount of variance while spatial extent of the sampling units (floras) explained a high amount of variance based on results from our variance partitioning analyses. This leads to the conclusion that spectral diversity must be carefully related to species diversity, explicitly taking into account potential area effects.