Conference Agenda

The extinction risk of species is globally monitored by the IUCN Red List (RL). Despite the goal to update extinction risk assessments at least every 10 years, over 21,000 assessments (~18% of the total) are considered outdated. The resources needed to keep assessments up to date present a clear sustainability challenge, and without a strategy to streamline the assessment process, the RL risks becoming unviable.

By bringing together RL practitioners and ecological modellers in a series of international workshops, we have designed and developed the sRedList platform, a user-friendly web-interface that guides assessors through a step-by-step procedure to assess species’ extinction risk. The platform returns all key parameters for RL assessment together with suggested RL categories its uncertainty.

Facilitating and streamlining the assessment procedure, the platform can reduce costs and time required to perform assessments, therefore improving our capacity to track global trends in the conservation status for multiple taxa. By eliminating technical barriers that prevent assessors from using certain data and analyses, the platform will also reduce taxonomic inconsistencies and the number of Data Deficient species. Overall, the sRedList platform has the potential to underpin the viability of the RL in the future.

Reliable data on different habitats’ extents and their changes are a much-needed resource for assessing species’ conservation-status under different IUCN Red List criteria, especially by supporting estimations of species’ area of habitat (AOH). We will present GlobES – global time-series for 58 terrestrial, freshwater, and intertidal ecosystem types conforming to the Red List’s habitat classification scheme. To achieve this thematic detail, but also sufficient mapping accuracy and consistency for reliable AOH and AOH-change estimations, we integrated quality-assured information derived from >40 satellite-based and in-situ datasets (covering land cover, land use, hydrology, climate, soil, coastal and stream topography, etc.). Comprehensive validation against millions of reference records show high overall accuracies and improved habitat representations in species’ ranges compared to existing products. The modular GlobES modelling framework is flexible regarding specific input layers, allowing for continued improvements, e.g., as unbiased time-series become available for more land-cover/use classes. The first version of the time-series will soon be published (open-access/FAIR). We will showcase how GlobES data can support improved AOH mapping, and, when integrated with different types of species-level information and covariates in sophisticated models, also support approximations of areas of occupancy for large species groups.

Hunting-induced defaunation poses an important challenge for biodiversity monitoring in tropical ecosystems, since it goes undetected by conventional remote sensing methods used for tracking deforestation. This information gap introduces biases in global conservation assessments of vertebrate species, potentially overestimating their distribution based solely on forest extent. To address this gap, we conducted a pantropical evaluation of hunting impacts on tropical bird and mammal abundance. We modelled hunting impacts using an extensive database of abundance estimates, predictors of hunting pressure, and biological traits that render species sensitive to hunting, while accounting for spatial and phylogenetic autocorrelation. We found that body mass, distance to hunter’s access points and travel time to urban markets were the most important predictors of hunting-induced declines of bird and mammal abundance. Then, we used our models to identify hotspots of defaunation at the pantropical scale for targeted conservation interventions. Lastly, we showcase how integrating our models in the sRedList platform will enhance Area of Habitat (AOH) maps by excluding suitable areas that are unlikely to be occupied. Additionally, it will enhance the application of IUCN criteria C and D by adjusting population size estimates based on predicted densities.

Generation length is defined as the average age of parents of the individuals in the population, and is a key parameter to assess species’ extinction risk using IUCN Red List criteria. Generation length is needed to define a universally comparable time horizon (either past or future) across different taxa over which species’ decline is measured. Yet, the information on generation length is still largely missing, and even among terrestrial vertebrates, which are fully assessed in the Red List, generation length is comprehensively available only for birds and mammals. This lack of knowledge inevitably affects the applicability of Red List criteria. Here, we used Generalized Additive Models to predict the generation lengths for squamates and amphibians based on a set of available data such as morphological traits, reproductive traits, phylogeny and climate. We found generation length increased with the size of the species and decreased with warmer climate for both groups, with snakes and Asiatic salamanders having the longest generation length on average. Our predictions can be used in future Red List assessments, expanding the applicable Red List criteria to assess past as well as future projected declines due to climate change.

Despite notable successful efforts in conservation policies, there is a clear global trend of declining biodiversity. Part of this is due to the crippling lack of information and man-power available to conservationists, straining efforts for all but the most popular (mostly vertebrate) species. This is reflected in the lack of extinction risk assessments for the neglected majority. To overcome data and analytical limitations, major current efforts are being made using big data and machine learning. Here we present project Asterisk, which attempts to automate much of the process of data collection and analysis to reach preliminary extinction risk assessments for invertebrates. The workflow is composed of multiple interrelated projects, including a global threat GIS database, automated extraction of location data from unstructured text, and extinction risk prediction using minimal data on species distributions and the threats facing them. The full workflow is made openly available through multiple online tools and R packages in constant development and update. Our goal is to multiply the pace of extinction risk assessments for the millions of species still lacking them and this way provide the necessary tools for the better conservation and management of biodiversity across the world.

The flowering plants (Angiosperms) are a large clade of ~330,000 species. Despite global and regional efforts over recent decades, the extinction risk of most (~70%) of these species remains unknown. We address this shortfall in knowledge by using the World Checklist of Vascular Plants to generate the first comprehensive set of predictions for all angiosperms (flowering plants).

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 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.