Skip to content

AI crop modelling

Recognised as the most water efficient, cotton-producing country in the world, Australia is three times more efficient than the global average. While this is commendable, water remains the most valuable resource to a cotton farmer and opportunities to further increase water efficiency are always being sought.

While current practices ensure that farmers are water efficient, looming issues around climate change will have a significant impact on temperature variability and rainfall, which may in turn make water an even scarcer resource. 

Lack of water can lead to reduced yields and boll sizes, making the crop less financially-viable for farmers and potentially threatening the overall production totals.

With 60% of the cotton produced being used for clothing items, a decrease in cotton production doesn’t only impact the farmers and the agricultural industry, it could also mean half of your wardrobe.

Surface irrigation systems, where the field itself is utilised to distribute the water, currently represent around 90% of irrigated cotton in Australia, but have typical water use efficiencies of 50-80%. Soil-water measurements are vital to optimise the water use efficiency of these types of systems.

Australia’s cotton growers produce enough cotton to clothe 500 million people each year.

UniSQ researchers leading the way

Current industry-standard soil-water sensors require contact with the soil and can only provide the information based on a single point in the field. While this enables farmers to have some oversight of soil-water conditions, it does not account for the spatial variability found in the entire field. 

UniSQ researchers have developed an Artificial Intelligence crop model that can determine current and predict future daily soil-water, nitrogen and fruit load of cotton plants based on the day of the season, weather data, soil electric conductivity, soil moisture, vegetation indices, plant density and canopy size, and visual plant response captured using cameras.

Flow on effects

These improvements to testing allow for a more over-arching view of the current water-soil conditions in any given field.

Using this technology enables farmers to achieve potential savings by optimising water and fertiliser management, water and fertiliser use efficiency, and crop productivity. The flow on effect is huge – with the potential to apply these models to any crop and the ability to be developed during crop season.


Be part of industry transformation

Further your studies and passion for research with UniSQ’s Higher Degrees by Research.