Australia's national science agency, CSIRO, in partnership with Meat & Livestock Australia (MLA) and Google Australia, has launched a global competition with a US$75,000 prize pool to advance the use of artificial intelligence (AI) in agriculture.
Hosted on Kaggle, the challenge seeks to improve the accuracy and efficiency of estimating pasture biomass. This is the amount of grass and other edible plants available for livestock to graze, a critical factor in grazing management that impacts productivity, environmental sustainability and biodiversity.
Grazing systems cover around half of Australia's landmass and roughly a quarter of the Earth's land surface. Accurately measuring pasture biomass enables farmers to balance livestock needs with pasture regrowth, supporting better grazing decisions and long-term land health.
Competitors will use pasture images linked with detailed field measurements to train an AI model to estimate pasture availability for livestock.
CSIRO Senior Principal Research Scientist Dr Dadong Wang said the images capture diversity across seasons, geographic locations and pasture species compositions.
"Each image is paired with detailed measurements, such as how tall the plants are and how green and healthy they look, based on how the pasture reflects light," Dr Wang said.
"By combining images with field data, we've collated a dataset that allows AI models to learn in more than one way.
"It can estimate pasture biomass directly from the images or combine the images with plant health and vigour information to produce even more accurate results."
Participants will use images to predict the amount of pasture available and the quantity of other plant species, like clover, with greater accuracy and usability than current approaches.
This could reduce the need for manual sampling, provide farmers with faster and more reliable information to guide grazing decisions, and shape the next generation of digital pasture measurement tools.
MLA's Group Manager – Science and Innovation Mr Michael Lee said the prospect of new methods to improve pasture biomass estimation was exciting.
"If successful, an AI-powered, machine vision approach will reduce the time and cost associated with manual sampling," Mr Lee said.
"The rapid and accurate differentiation of biomass components also feeds into improving producer's ability to both determine current pasture quality, as well as predicting quantity and quality into the future."
Google Australia's Partnerships Principal Mr Scott Riddle said the project shows how research and industry can work together to address real-world challenges.
"As part of the Digital Future Initiative, we're delighted to partner with CSIRO and MLA to support the future of Australian farming – with a bit of help from AI," Mr Riddle said.
"By connecting CSIRO's deep scientific expertise and MLA's industry knowledge with the 26 million innovators on Kaggle, we're putting a global AI community to work to try and find new approaches to a longstanding problem."
Insights from this competition could form the basis of future digital tools, supporting more sustainable grazing systems in Australia and worldwide.
Explore the challenge and submit your entry here before it closes on 28 January 2026.
This work has been supported by FrontierSI (previously known as the Cooperative Research Centre for Spatial Information).
