Open Agrobiodiversity Accounting Kit (OAAK)

BY Genomic Gastronomy & ​​NICETRAILS

VIDEO

Concept & Prototype description

OAAK is a digital tool designed to facilitate participatory, place-based species quests for agricultural biodiversity accounting on small-scale farms. It connects eaters and farmers by generating meaningful data to meet biodiversity standards or participate in emerging biodiversity credit schemes. The tool integrates AI and citizen science to assess and monitor agricultural biodiversity in Europe, supporting the agricultural green transition. OAAK works with agritourism agencies, where farmers register their farm details (location, crops, dates) and offer farm visits through the app. Visitors are then guided through specific quests that focus on cataloging species (plants, animals, fungi) and tracking biodiversity changes over time, enhanced by regional data and continuous assessments.

Connection to the scenario
Future scenario: Patterns that Persist

We started with the scenario “Biodiversity as the measure of healthy food systems” and the key assumption that top-down regulation or a biodiversity credits market will implore or encourage farmers to measure biodiversity. Small-scale farms have less resources to do this than large-scale conventional farms do, and farmers are already very busy and overwhelmed with paperwork. The task of measuring biodiversity needs to be simplified.
After identifying the values we held like pleasure, optimism, openness, resilience, adaptability, commitment, and equity, we asked ourselves the critical question: “What future are we building?” When we look at the (agro-)tech market, we see the development of smart machines, and the de-skilling of humans. Humans are more and more isolated by alienating technologies. In contrast to this, OAAK aims to build a future that is: convivial, community-oriented, re-skilled, healthy, and resilient. Thus, we want to build a technology that is: social, embedded, multi-purpose, and cooperative.

Technology

OAAK’s TLR-5 prototype will leverage several technologies to facilitate species identification and biodiversity tracking. It will use AI vision models for species detection, potentially incorporating multi-modal large language models (LLMs) for image recognition. LLMs will help create a dynamic guide based on user inputs, location, and farmer expectations. Data APIs, like the iNaturalist API, will gather contextual data, while AI embeddings will help visualize and compare the collected data in an intuitive way. The tool will guide users to discover species through a conversational interface, linking specific species to their environments. Ultimately, OAAK will compile a crowdsourced agrobiodiversity dataset, incorporating species, locations, and times.