Implementation of automated biodiversity monitoring lags behind its potential

Abstract Efforts to conserve and restore biodiversity rely on widespread, accurate, and timely monitoring to understand where and how fast biodiversity is changing and develop appropriate responses. The current landscape of biodiversity assessments typically requires time-intensive and costly field surveys that rarely match the pace of monitoring needs. Automated biodiversity monitoring could increase the speed and accuracy at which information on species and ecosystems is gathered, and thus improve the data used in conservation decision making. However, the extent of existing monitoring with these ‘digital assets’ is not well characterized, limiting how and where current data can be used and our understanding of where key gaps remain. Here, we catalog existing digital assets for monitoring living organisms globally, finding that they predominantly come from satellite data (68%) and monitor plants at coarse taxonomic resolutions. Additionally, the spatial coverage is biased towards regions in North America and Europe, and data availability frequently lags well behind its collection. We also developed a framework to describe key steps in creating digital assets that can be used to identify bottlenecks in their production, and suggest strategies to address barriers. In particular, investing in developing data processing algorithms, building and automating workflows, and increasing accessibility of existing data could help rapidly increase the availability of digital assets for biodiversity monitoring. Accelerating and expanding development of automated biodiversity assessments should pay broad dividends for managing and conserving biodiversity globally.