Skip to content
menu icon

‘Tech’ efforts bridge the gap between science and user-friendly products

On-farm decisions could be made more certain with improved real-time predictions of stored soil water. The SoilWaterNow project aims for growers to have access to accurate, rapid spatial and temporal plant available water measures by the end of 2026.
Photo: Nicole Baxter

Agricultural technology companies will soon be able to more easily test the latest plant-available water (PAW) models, developed by SoilWaterNow researchers, to overcome commercial risks associated with ‘productising’ researcher-developed technology.

The SoilWaterNow project leader, University of Sydney Professor Tom Bishop, says the research team has built a scalable approach to model soil PAW trends for the whole soil profile, and at the sub-paddock scale, using multiple layers of input data from on-farm sources and off-farm sources.

Professor Bishop says agtech companies need to test models with their users before investing the time and money required to ‘productise’ the science underpinning them.

“The challenge is in facilitating feedback with agtech companies, helping them to integrate the science behind the models and deliver accurate PAW data to growers and agronomists,” says Professor Bishop.

“Agtech companies are key to taking the latest science and developing it into valuable and user-friendly information products that will be widely adopted. That development cost is often underestimated, and so we need to create bridges that enable companies to provide great information products on PAW to agronomists and growers.”

Low-cost link

To mitigate these commercial risks, an application programming interface (API) will be used. This is a low-cost mechanism that creates a link between PAW models that have been developed and their evaluation. It means that as researchers develop PAW models, they can automatically engage agtech companies via the API to produce PAW data for any location and point in time using newly developed PAW models. “The process provides feedback on models across a diverse set of agronomist and grower users. This can then be fed back to researchers to improve the model or the inputs to better meet the need on-farm,” Professor Bishop says.

“Evaluation usage also provides us, as the researchers, with user feedback on model performance. This feedback ensures future research and development activities are oriented towards delivering user needs.”

Professor Bishop says this approach also helps when the project ends. “When researchers stop maintaining the API, agtech companies will be in a great position to understand the costs and benefits of integrating PAW models into their platforms based on the identified needs of their agronomists and grower users.”

Model adoption

This leads to overcoming the challenges of turning the science behind the models into actual information products that aid decision-making and improve grower profit.

“There’s a lot of work associated with developing, testing and commercialising an information product that has the look, feel, and usability to become widely adopted. Agtech companies are well-positioned to do that work. What we’re doing is making it easier to access the latest science that can be used to power the smarts behind those products.”

This includes making the mathematical models developed by researchers easier to scale and operate. This is a specialist’s role and a data science software engineer from the Sydney University Informatics Hub has been called in and already reduced computational time and cost by 50 per cent.

“The engineer rewrote the base PAW model into a more scalable and widely used programming language and used coding techniques to make the model run efficiently at scale.

“Work like this should make these new PAW models more adoptable and help ensure the research outputs can also be leveraged in other research and development projects.”

The project aims for growers to have access to accurate and rapid spatial and temporal measures of PAW by the end of 2026, but this work is speeding up that timeframe and the delivery of benefits on-farm.

‘Soil water nowcasting for the grains industry’ is led by the University of Sydney in a collaboration with CSIRO, the University of Southern Queensland, Australian National University and the Bureau of Meteorology.

More information: Tom Bishop, 0405 023 457,

back to top