How well does the plant fossil record capture vegetation phylogenetic diversity?

Reconstructing past biodiversity changes, and integrating these with modern biodiversity assessments, requires that fossil assemblages accurately capture key components of diversity (as represented by biodiversity metrics, for example). This is particularly challenging for the plant fossil record, where separate organs such as sporomorphs (pollen and spores) and leaves must be used as proxies for vegetation composition and diversity. Although much attention has been focused on how well fossil plant assemblages capture variations in species richness, other aspects of diversity have until recently been relatively overlooked.

Here, I focus on phylogenetic diversity (PD), which represents the amount of evolutionary history contained in an assemblage of taxa. It can therefore provide a more detailed assessment of biodiversity gains and losses through time and space, and their underlying causes and consequences, relative to simple counts of the number of species present in a sample. To date, however, PD has been underexplored by palaeoecologists, and it is not currently known how well variations in vegetation PD across broad spatial scales are captured by sporomorph assemblage data. I compare estimates of seed plant PD from vegetation data and surface pollen samples from across North and South America. The results indicate a relatively low concordance between vegetation and pollen PD, and differing relationships with climate data, suggesting that sporomorph data cannot be used as a straightforward PD record. Other data sources (e.g. aDNA data for late Quaternary datasets, macrofossil data in deeper time settings) need to be considered for reconstructing vegetation PD through time.