Changing the game in cotton nitrogen management

Cotton growers in Australia’s Gwydir/Namoi region are constantly looking for ways they can innovate to maintain high yields while reducing water and nutrient costs. With Nitrogen both a major input cost and key determinant of yield performance, much of that innovation is focused on knowing at the earliest possible stage the exact amount of Nitrogen to apply that will result in the best possible yield.

During the 2017/18 growing season, the region’s leading agronomists and growers, including Auscott (a division of J.G. Boswell), Landmark and McGregor Gourlay, partnered with FluroSat, an Australian crop health analytics startup, to trial remote sensing crop analysis on over 14,000 acres of cotton being grown on 71 irrigated and dryland farms.

Using a UAV equipped with the MicaSense series RedEdge, FluroSat flew over the farms during critical growth stages. The RedEdge sensor was chosen because it includes the red edge band, which is less volatile than the near infrared in response to changes in plant moisture, resulting in more precise detection and measurement of chlorophyll.

The data collected was then calibrated by conducting tissue sampling in each field in coordination with each RedEdge flight. The calibrated maps reduce and even remove the need to tissue sample for the rest of the season and subsequent seasons.

Using the FluroSat analytics platform, the analysis was then presented in maps that display the plant Nitrogen concentration and their sufficiency levels at different dates across the season. These values can be averaged across the field or narrowed to individual rows for trialing different growing strategies or varieties. Similar Nitrogen values are automatically grouped into management zones that can be used for Variable Rate (VR) application.

The agronomists in the trial also used the FluroSat platform to analyze the Nitrogen maps against other layers like NDVI and water stress to have a wider picture of the factors contributing to the current Nitrogen uptake. The insights they gain using FluroSat and RedEdge resulted in more timely and precise in-crop fertilization recommendations that ensured that Nitrogen was being used only where it can make a positive impact on yield, reducing fertilizer costs by 30–35% on average. At the end of the season, yield data from a number of monitored fields was compared against FluroSat’s analysis, showing good correlation coefficients in the range of 0.4–0.5 from as early as mid-December.

Conclusion

Improving Nitrogen management is a key factor in lifting Australia’s already impressive productivity to even higher levels. The type of targeted Nitrogen analysis that was done in this study can be more effective than the commonly used NDVI index as it goes beyond simply correlating Nitrogen with crop biomass, by filtering out all changes unrelated to Nitrogen, remaining useful through the season, unaffected by saturation, resulting in an accurate and reliable quantification of the Nitrogen “fitness” of the plant matter. By applying the insights gained using RedEdge and FluroSat, growers are now able to intervene earlier and with more precision to lift average yield across their fields by 20%.