Presentation: AI-driven tool for invasive species data search and integration

Authors

Samuel Vander Velpen (Victoria University of Wellington, Nelson AI, Sequench), Mengjie Zhang (Victoria University of Wellington), Bing Cue (Victoria University of Wellington), Julian Maclaren (NAI), Anastasija Zaiko (Sequench)

About

Global proliferation of invasive marine species poses a significant threat to aquatic ecosystems, disrupting ecological processes and impacting biodiversity, resource utilization, and human activities dependent on these environments. In response, the EU initiated GuardIAS, a collaborative project leveraging advanced technologies such as AI, environmental DNA, satellite imagery, and robotics to detect and mitigate spread of invasive species.

A primary challenge within GuardIAS is the lack of coherence in invasive species data, characterized by inconsistencies across databases and limited access to usable information. To address this, we are developing an interactive Python-based web tool that features an AI trained to query other AI tools. This web tool will facilitate the setup of queries to gather, extract, and format information on invasive species.

Relevant information, as specified in the query or determined by AI training, will focus on data critical for monitoring and managing invasive aquatic species, including ideal living conditions, migration methods, reproduction rates, and known prevention strategies. The extracted data will be delivered in formats such as CSV and GeoTIFF and uploaded to GBIF and EASIN to create a more uniform database.

Download: How Artificial intelligence can help in marine biosecurity-1