Estimating regional species richness: The case of China's vascular plant species

Abstract

AbstractAimThe estimation of regional species richness has been a major challenge in ecology but is crucial for setting up conservation priorities. Discovery curves are a principal tool for estimating regional richness, but they have been criticized for being too sensitive to historical fluctuations in species discovery. In this study we propose a new discovery model that considers historical influences. We have applied the model to estimate the number of vascular plant species in China, a country of mega‐biodiversity that has also suffered endless wars and social turmoil in its modern history.LocationChina.Time period1755–2000.Major taxa studiedVascular plant species.MethodsWe compiled the discovery time for each vascular species from the complete volumes of Flora of China, leading to 31,220 valid species names. We applied our model to these data and compared the performance of our model with existing discovery models and two other methods (species–area relationships and a taxonomic rank curve model). We also tested our model with three other independent datasets and one simulation study.ResultsOur new method estimated there to be 36,554 (± 2,708) vascular plant species in China. Our model accounted very well for the effect of historical events and was robust to different periods of data that were used to estimate the total richness. Our model outperformed all other models that were compared. We found that 5,334 species remained to be discovered in China and it would take about 50 years to discover all the species should the current discovery rate of 110 species per year persist.Main conclusionsSpecies discovery curves, with historical effects being properly accounted for, offer a promising tool for estimating regional species richness. Our model is robust to the effect of historical events and provides by far the most accurate and reliable estimates of species richness of test data, including vascular plant richness in China.