10.5061/DRYAD.V7M14
FitzJohn, Richard G.
University of British Columbia
Macquarie University
Pennell, Matt W.
VU University Amsterdam
Zanne, Amy E.
Missouri Botanical Garden
George Washington University
Stevens, Peter F.
University of Missouri
Tank, David C.
University of Idaho
Cornwell, William K.
VU University Amsterdam
UNSW Sydney
Pennell, Matthew W.
University of Idaho
National Evolutionary Synthesis Center
Data from: How much of the world is woody?
Dryad
dataset
2015
Herbaceousness
woodiness
sampling bias
2015-04-08T00:00:00Z
2015-04-08T00:00:00Z
en
https://doi.org/10.1111/1365-2745.12260
1575385 bytes
2
CC0 1.0 Universal (CC0 1.0) Public Domain Dedication
1.The question posed by the title of this paper is a basic one, and it is
surprising that the answer is not known. Recently assembled trait datasets
provide an opportunity to address this, but scaling these datasets to the
global scale is challenging because of sampling bias. Although we
currently know the growth form of tens of thousands of species, these data
are not a random sample of global diversity; some clades are exhaustively
characterised, while others we know little–to–nothing about. 2.Starting
with a database of woodiness for 39,313 species of vascular plants (12% of
taxonomically resolved species, 59% of which were woody), we estimated the
status of the remaining taxonomically resolved species by randomisation.
To compare the results of our method to conventional wisdom, we informally
surveyed a broad community of biologists. No consensus answer to the
question existed, with estimates ranging from 1% to 90% (mean: 31.7%).
3.After accounting for sampling bias, we estimated the proportion of
woodiness among the world's vascular plants to be between 45% and
48%. This was much lower than a simple mean of our dataset and much higher
than the conventional wisdom. 4.Synthesis: Alongside an understanding of
global taxonomic diversity (i.e., number of species globally), building a
functional understanding of global diversity is an important emerging
research direction. This approach represents a novel way to account for
sampling bias in functional trait datasets and to answer basic questions
about functional diversity at a global scale
Results-woodinessAll results from the analyses conducted in the paper and
results from our survey of researcherswood-supporting.tar.gz