GLOBAL MULTIMODAL APPROACH TO BIODIVERSITY ASSESSMENT

Satellite based monitoring for biodiversity assessment combined with local measurements of biodiversity, through field measurements and manual observations have enormous potential to provide an end-to-end solution to quantify the main drivers of biodiversity enrichment at global scale. In this work, we propose an end-to-end solution for a scalable and multimodal foundation model for biodiversity assessment and we explore the utility of biodiversity indices calculated on above ground biomass (AGB) maps predicted by machine learning models for the Karukinka Natural Park in Chile and the Maasai Mau Forest Reserve in Kenya over the years.