Australia’s national science agency, CSIRO, in collaboration with Meat & Livestock Australia (MLA) and Google Australia, has initiated a global competition aimed at harnessing artificial intelligence (AI) to improve pasture management. With a prize pool of US$75,000, the challenge, hosted on the platform Kaggle, seeks innovative solutions to enhance the accuracy and efficiency of estimating pasture biomass, which is crucial for livestock grazing.
Pasture biomass refers to the amount of grass and other edible plants available for grazing. This measurement plays a vital role in grazing management, influencing productivity, environmental sustainability, and biodiversity. Grazing systems cover approximately half of Australia’s landmass and about a quarter of the Earth’s land surface. By accurately measuring pasture biomass, farmers can better align livestock needs with pasture regrowth, which supports sustainable grazing practices and long-term land health.
Participants in the competition will utilize images of pastures, supplemented by comprehensive field measurements, to train AI models that estimate pasture availability. Dr Dadong Wang, Senior Principal Research Scientist at CSIRO, emphasized the diversity of the dataset. The images capture variations across seasons, geographical locations, and species compositions. “Each image is paired with detailed measurements, such as the height of the plants and their health based on light reflection,” Dr Wang explained.
This dataset allows for multiple learning approaches. AI models can either estimate pasture biomass directly from the images or enhance accuracy by integrating plant health and vigor information. The innovative use of AI has the potential to significantly reduce the need for manual sampling, providing farmers with quicker and more reliable data to inform grazing decisions.
Mr Michael Lee, Group Manager – Science and Innovation at MLA, expressed enthusiasm about the possibilities offered by the challenge. “If successful, an AI-powered machine vision approach will minimize the time and costs associated with manual sampling,” Mr Lee stated. He noted that the ability to rapidly and accurately differentiate biomass components could enhance producers’ understanding of current pasture quality and improve predictions of future biomass.
On behalf of Google Australia, Mr Scott Riddle, Partnerships Principal, highlighted the collaboration’s significance. “As part of the Digital Future Initiative, we are pleased to partner with CSIRO and MLA to advance the future of Australian farming using AI,” Mr Riddle remarked. He noted that the collaboration leverages CSIRO’s scientific expertise and MLA’s industry insights, connecting them with the global community of innovators on Kaggle to find new solutions to longstanding challenges.
The insights gained from this competition could pave the way for the development of future digital tools that promote sustainable grazing systems in Australia and across the globe. Interested participants can explore the challenge and submit their entries until January 28, 2026. This initiative has received support from FrontierSI, previously known as the Cooperative Research Centre for Spatial Information.
