Summary

By interrogating yield maps over many years, we know that rice yield can vary by at least 50%, and sometimes over 100%, in every single paddock. Even 3-4t/ha variation can cost growers over $1000/ha in lost production, and the industry over $150M annually. Keeping in mind that the same amount of water and nutrients are applied on average across each paddock, leading to massive variation in WUE (Water Use Efficiency). What’s holding the industry back is whilst some of the base Research (‘R’) has been done, there is a significant amount of Development (‘D’) work required to convert the ‘algorithms’ into practical tools deployed for on-ground adoption by the majority of producers. This project will significantly close the gap between research and adoption. Precision agriculture (PA) tools offer the ability to solve problems such as providing simple and automated tools for variable rate aerial application, pest and disease anomaly detection, harvest timing, and soil constraints; but they are out of reach for most rice growers and agronomists.  The research involved in some of these applications (such as utilising satellite imagery to determine nitrogen rates and the presence of specific pests and diseases), is fraught with challenges because the spectral bands are different with each satellite/drone, the reflectance values change on each satellite/drone pass, and with different varieties, geographical locations and slopes, which means, that one model will never likely fit all cases.    It is the reason why the agronomist will always play a vital role in ground truthing and determining the rates to apply to a crop, and why agronomists need to confirm what pest/weed/disease/disorder is present and the type of action required.
 
The project team will work closely with agronomists, farmers, NSW government, harvest contractors, SunRice, and technology suppliers to deliver four (4) simple, effective, and semi-automated PA tools for the rice industry to use; which include:

1. Rapid soil mapping for ameliorating soil constraints 
2. In-crop anomaly detection using satellite imagery
3. Automated variable rate aerial application using satellite imagery
4. Predicting harvest timing (sub-paddock level) using satellite imagery and IoT sensors

We expect 75% of rice farmers and agronomists utilising one of more of these tools by 2025. Each tool covers a part of the growing cycle, and provides farmers and agronomists a way to get the information they need without having to utilise complex software or technology or concern themselves with file formats.  The tools enable them to view the information, add their expertise to recommendations, and then act.  Rice is slightly different to other crops – the demands of the crop towards the end of December mean that farmers and agronomists have very little time to do precision ag activities, and hence they are often of lower priority – hence why it is SO IMPORTANT to build quick a

Program

Rice

Research Organisation

Data Farming Pty Ltd