Extinction risk predictions for the world's flowering plants to support their conservation.

More than 70% of all vascular plants lack conservation status assessments. We aimed to address this shortfall in knowledge of species extinction risk by using the World Checklist of Vascular Plants to generate the first comprehensive set of predictions for a large clade: angiosperms (flowering plants, c. 330 000 species). We used Bayesian Additive Regression Trees (BART) to predict the extinction risk of all angiosperms using predictors relating to range size, human footprint, climate, and evolutionary history and applied a novel approach to estimate uncertainty of individual species-level predictions. From our model predictions, we estimate 45.1% of angiosperm species are potentially threatened with a lower bound of 44.5% and upper bound of 45.7%. Our species-level predictions, with associated uncertainty estimates, do not replace full global, or regional Red List assessments, but can be used to prioritise predicted threatened species for full Red List assessment and fast-track predicted non-threatened species for Least Concern assessments. Our predictions and uncertainty estimates can also guide fieldwork, inform systematic conservation planning and support global plant conservation efforts and targets.

Evidence-based guidelines for automated conservation assessments of plant species.

Assessing species' extinction risk is vital to setting conservation priorities. However, assessment endeavors, such as those used to produce the IUCN Red List of Threatened Species, have significant gaps in taxonomic coverage. Automated assessment (AA) methods are gaining popularity to fill these gaps. Choices made in developing, using, and reporting results of AA methods could hinder their successful adoption or lead to poor allocation of conservation resources. We explored how choice of data cleaning type and level, taxonomic group, training sample, and automation method affect performance of threat status predictions for plant species. We used occurrences from the Global Biodiversity Information Facility (GBIF) to generate assessments for species in 3 taxonomic groups based on 6 different occurrence-based AA methods. We measured each method's performance and coverage following increasingly stringent occurrence cleaning. Automatically cleaned data from GBIF performed comparably to occurrence records cleaned manually by experts. However, all types of data cleaning limited the coverage of AAs. Overall, machine-learning-based methods performed well across taxa, even with minimal data cleaning. Results suggest a machine-learning-based method applied to minimally cleaned data offers the best compromise between performance and species coverage. However, optimal data cleaning, training sample, and automation methods depend on the study group, intended applications, and expertise.

La valoración del riesgo de extinción de las especies es vital para el establecimiento de prioridades de conservación. Sin embargo, los esfuerzos de valoración, como los que se usan para generar la Lista Roja de Especies Amenazadas de la UICN, tienen brechas importantes en la cobertura taxonómica. Los métodos de valoración automatizada (VA) están ganando popularidad como reductores de estas brechas. Las elecciones realizadas en el desarrollo, uso y reporte de resultados de los métodos de VA podrían obstaculizar su adopción exitosa o derivar en una asignación deficiente de recursos para la conservación. Exploramos cómo la selección del tipo de limpieza de datos y el nivel, grupo taxonómico, muestra de entrenamiento y el método de automatización afectan el desempeño de las predicciones del estado de amenaza de las especies de plantas. Usamos los registros de la Global Biodiversity Information Facility (GBIF) para generar las valoraciones de las especies de tres grupos taxonómicos con base en seis métodos diferentes de VA basados en la presencia de las especies. Medimos el desempeño de cada método y cobertura después de una limpieza de presencia cada vez más estricta. La información de la GBIF limpiada automáticamente tuvo un desempeño comparable con los registros de presencia limpiados manualmente por expertos. Sin embargo, todos los tipos de limpieza de datos limitaron la cobertura de las valoraciones automatizadas. En general, los métodos basados en el aprendizaje automático tuvieron un buen desempeño en todos los taxones, incluso con una limpieza mínima de datos. Los resultados sugieren que un método basado en el aprendizaje automático aplicado a información con la mínima limpieza ofrece el mejor equilibrio entre el desempeño y la cobertura de la especie. A pesar de esto, la limpieza óptima de datos, la muestra de entrenamiento y los métodos de automatización dependen del grupo de estudio, las aplicaciones deseadas y la experiencia.

Predicting extinction risk of Brazilian Atlantic forest angiosperms.

Understanding how plant life history affects species vulnerability to anthropogenic disturbances and environmental change is a major ecological challenge. We examined how vegetation type, growth form, and geographic range size relate to extinction risk throughout the Brazilian Atlantic Forest domain. We used a database containing species-level information of 6,929 angiosperms within 112 families and a molecular-based working phylogeny. We used decision trees, standard regression, and phylogenetic regression to explore the relationships between species attributes and extinction risk. We found a significant phylogenetic signal in extinction risk. Vegetation type, growth form, and geographic range size were related to species extinction risk, but the effect of growth form was not evident after phylogeny was controlled for. Species restricted to either rocky outcrops or scrub vegetation on sandy coastal plains exhibited the highest extinction risk among vegetation types, a finding that supports the hypothesis that species adapted to resource-limited environments are more vulnerable to extinction. Among growth forms, epiphytes were associated with the highest extinction risk in non-phylogenetic regression models, followed by trees, whereas shrubs and climbers were associated with lower extinction risk. However, the higher extinction risk of epiphytes was not significant after correcting for phylogenetic relatedness. Our findings provide new indicators of extinction risk and insights into the mechanisms governing plant vulnerability to extinction in a highly diverse flora where human disturbances are both frequent and widespread.