In the heart of Italy, researchers at the Department of Agricultural and Food Sciences (DISTAL) at the University of Bologna, led by Vito Aurelio Cerasola, have been making waves in the world of precision agriculture. Their latest study, published in Smart Agricultural Technology, delves into the intricacies of hyperspectral imaging to optimize nitrogen management in processing tomatoes. This isn’t just about growing better tomatoes; it’s about revolutionizing how we approach agriculture in the face of climate change and resource scarcity, with far-reaching implications for the energy sector.
Precision nitrogen management is a critical component of sustainable agriculture. Too much nitrogen can lead to environmental degradation, while too little can stifle crop growth. The challenge lies in finding that sweet spot— the optimal nitrogen rate—that maximizes yield without compromising the environment. This is where hyperspectral imaging comes into play.
Cerasola and his team explored two methodological approaches to estimate the optimal nitrogen rate using hyperspectral imaging. The first approach, dubbed the ‘N uptake approach,’ focuses on virtually reproducing the critical N uptake curve. This involves estimating both aboveground biomass and crop nitrogen uptake. The optimal nitrogen rate is then computed as the difference between the critical N uptake and the actual N uptake.
The second approach centers around monitoring the Nitrogen Nutrition Index (NNI) and biomass. Here, the biomass is used to calculate the critical N uptake, which, when combined with the estimated NNI, helps retrieve the actual crop nitrogen uptake.
To achieve this, the researchers employed an unmanned aerial vehicle equipped with hyperspectral sensors. They collected canopy reflectance data across the full spectrum (400–1000 nm) at five growth stages of processing tomatoes grown under different nitrogen rates. The data was then fed into three nonparametric algorithms: Gaussian Process Regression (GPR), Support Vector Regression (SVR), and Partial Least Square Regression (PLSR).
Cerasola explains, “The pivotal role of biomass in the selected nitrogen rate estimation approaches led us to explore two distinct biomass estimation methods. The direct biomass retrieval from spectral data was compared with the indirect biomass retrieval from the remotely sensed Leaf Area Index (LAI) applying empirical regressions.”
The results were promising. PLSR outperformed the other algorithms in estimating nitrogen uptake, while SVR excelled in estimating NNI and direct biomass. Interestingly, the indirect estimation of biomass outperformed the direct approach when GPR was used, although early growth stages posed challenges due to soil background interference.
Cerasola notes, “The NNI approach outperformed the N uptake approach in estimating the optimal nitrogen rate, especially when the biomass is directly retrieved from GPR. The promising estimation performances in nitrogen rate estimation revealed the effectiveness of hyperspectral imaging in entering the agronomical scheduling of precision nitrogen management.”
So, what does this mean for the energy sector? As the world shifts towards renewable energy, the demand for sustainable agriculture practices grows. Efficient nitrogen management not only reduces environmental impact but also enhances crop productivity, ensuring a stable food supply. This, in turn, supports bioenergy initiatives, where crops are used as a renewable energy source.
This research isn’t just about growing better tomatoes; it’s about laying the groundwork for a more sustainable future. By harnessing the power of hyperspectral imaging and machine learning, we can optimize nitrogen management, reduce environmental impact, and enhance crop productivity. As we look to the future, the integration of these technologies into mainstream agricultural practices could revolutionize how we approach food and energy security. This study, published in ‘Intelligent Agricultural Technology’, is a significant step forward in this exciting journey.