Published February 1, 2024 | Version 1.0
Dataset Open

Predicted extinction risk for all Angiosperm (flowering plant) species

  • 1. Royal Botanic Gardens Kew
  • 2. ROR icon Royal Botanic Gardens, Kew

Description

Species level predictions of extinction risk for all Angiosperm (flowering plant) species with uncertainty estimates. 

Note that these predictions may differ from published assessments. Published assessments, when available, should be used instead of predictions.

Field headings:

plant_name_id = plant name identifier from World Checklist of Vascular Plants

threatened = mean probability of being threatened

.lower = lower bound of 95% credible interval

.upper = upper bound of 95% credible interval

.width = width of the credible interval

.point = summary function (mean)

.interval function = highest-density interval mean

predicted_threat = predicted extinction risk class (threatened or not threatened)

confidence = confidence in the predicted class (low_confidence or confident)    

category = observed Red List category if assessed

observed_threat = extinction risk class for Red List assessed species (threatened or not threatened)

combined = combined observed and predicted threat class, observed takes precedence over predicted where there are both;

ipni_id = species identifier according to the International Plant Names Index (IPNI)

family = World Checklist of Vascular Plants (WCVP) family name

genus = World Checklist of Vascular Plants (WCVP) genus name

species = World Checklist of Vascular Plants (WCVP) species

taxon_name = binomial

A note on percentage threatened – users will note that deriving the percentage threatened from this dataset differs from the overall percentage threatened presented in the associated paper. This is due to the two different approaches used – see methods section in the associated paper. The first looks at the posterior predictive, a series of 0/1 (not threatened/threatened) predictions for each species across multiple (1000) draws. From each draw (one per species) a mean percentage is derived and this is averaged across all draws to obtain variation in that mean (95% credible interval). The second method is based on the posterior probability of being threatened i.e. whether the mean probability (from 1000 draws) was above our model threshold. In addition, we used the 95% credible interval of the probabilities to assign a value of confidence to each species. When the 95% credible intervals were completely below (not threatened), or completely above (threatened) the threshold, they were deemed confident, and where the credible intervals crossed the threshold, they were deemed not confident.

Future versions of this dataset will be released as the Red List and World Checklist of Vascular Plants datasets are updated and the model is rerun. Please cite the latest version and the associated paper:

Extinction risk predictions for the world's flowering plants to support their conservation (2024). Bachman, S.P., Brown, M.J.M., Leão, T.C.C., Lughadha, E.N., Walker, B.E. New Phytologist. https://nph.onlinelibrary.wiley.com/doi/full/10.1111/nph.19592

Files

Angiosperm_extinction_risk_predictions_v1.csv

Files (55.3 MB)

Name Size Download all
md5:e6fbee92fe19b783345cda27fc18eddc
55.3 MB Preview Download

Additional details

Related works

Dates

Available
2024-02-01