GeoPl@ntNet: A Platform for Exploring Essential Biodiversity Variables

TL;DR

The paper tackles turning Essential Biodiversity Variables into accessible, high-resolution maps across Europe. It introduces GeoPl@ntNet, a multimodal cascading pipeline combining a Deep-SDM for species and Pl@ntBERT for habitats, plus conformal-prediction-based indicators. Using Europe-wide data (5M presence-only + 90K presence-absence records) at 50 m resolution, it produces maps for over 10,000 species, 200+ habitat types, and seven biodiversity indicators, along with interactive ROI reports. The platform, delivered as a VueJS web app with WMS streams, aims to aid conservation planning and biodiversity monitoring, aligning with the EU biodiversity strategy for 2030.

Abstract

This paper describes GeoPl@ntNet, an interactive web application designed to make Essential Biodiversity Variables accessible and understandable to everyone through dynamic maps and fact sheets. Its core purpose is to allow users to explore high-resolution AI-generated maps of species distributions, habitat types, and biodiversity indicators across Europe. These maps, developed through a cascading pipeline involving convolutional neural networks and large language models, provide an intuitive yet information-rich interface to better understand biodiversity, with resolutions as precise as 50x50 meters. The website also enables exploration of specific regions, allowing users to select areas of interest on the map (e.g., urban green spaces, protected areas, or riverbanks) to view local species and their coverage. Additionally, GeoPl@ntNet generates comprehensive reports for selected regions, including insights into the number of protected species, invasive species, and endemic species.

Figures (5)

• Figure 1: Multimodal cascading pipeline to produce the maps. The Deep‐SDM (ResNet-6 ensemble) predicts species based on satellite, climatic, and environmental data, and the Deep-HDM (Pl@ntBERT encoder) predicts habitat types using the predicted species assemblages.
• Figure 1: Evaluation of the deep-SDM (with different metrics).
• Figure 2: Area of Interest report. Users select an area and see insights on threatened or invasive species, habitat status, and other metrics.
• Figure 3: Modules available. Home module (A): Offers an overview of key biodiversity metrics. Species module (B): Displays a ranked list of species predicted. Habitats module (C): Shows the most likely habitat types. Indicators module (D): Summarizes biodiversity indicators.
• Figure 4: Example maps. Top panel shows species distribution maps, middle panel shows biodiversity indicator maps, and bottom panel shows habitat suitability maps. They are available at a 50m resolution in Europe for over 10,000 species, over 200 habitats, and 7 indicators.