Main Content


We develop open-source software to curate and analyze large dataset in in biodiversity research and conservation (Github).


IUCNN is an open-source R package to predict the Red List status of species that have not yet been officially assessed for the IUCN global Red List. IUCNN uses Artificial Intelligence to predict the Red List status from species traits, geographic distributions, and remote sensing data using Deep learning. Using this approach, IUCNN can predict the Red List status for many thousands of species with over 80% accuracy within minutes. IUCNN is available as an R package and we describe the method in detail in this publication (Zizka et al., 2022).


A. Zizka

Bio-Dem is an open-source webapp to explore the link between biodiversity data availability and the socio-political situation in countries worldwide. Bio-Dem consists of three interactive graphs, and links the availability of species distribution data from the Global Biodiversity Information Facility with socio-political information from theVarieties of Democracy Institute. With Bio-dem, users can compare changes in the availability of distribution data and levels of democratization across the globe over the course of the 20th century. Using Bio-Dem, we showed that the availability of biodiversity data depends on the level of democratization and colonial history (Zizka et al., 2021). Bio-Dem won first prize in the 2021 Ebbe Nielssen Challenge.


Alexander Zizka
SampBias is a new method to measure and map the influence of varying accessibility of areas to biodiversity researchers on the collection intensity of distribution data. SampBias allows to compare the effect among individual types of infrastructure (e.g., settlements, roads, and rivers) and to compare the overall sampling bias between different datasets (e.g., from different taxa or regions). SampBias is implemented as an accessible open-source R package with detailed documentation and tutorial. We describe the method in detail in this publication (Zizka et al., 2021). SampBias won 2nd place in the 2016 Ebbe Nielsen Challenge.


Alexander Zizka
CoordinateCleaner is an R-package to detect erroneous geographic coordinates in large datasets on species distributions. CoordinateCleaner automatically detects a variety of typical georeferencing problems in biological databases, for example, invalid coordinates, coordinates in oceans, coordinates on country or province centroids, capitals and locations of biodiversity institutions, among others. CoordinateCleaner is open-source and part of the Ropensci software suite. Detailed documentation and tutorials and a scientific publication in which we describe the method are available (Zizka et al., 2019).