Developing a hybrid data-driven and informed model for prediction and mitigation of agricultural nitrous oxide flux hotspots

• Published in: Frontiers in environmental science Vol. 12
• Main Author: Vemuri, Nikhil
• Format: Journal Article
• Language: English
• Published: Lausanne Frontiers Research Foundation 22.03.2024; Frontiers Media S.A
• Subjects: Agricultural sciences; Anthropogenic factors; Approximation; Biogeochemistry; Climate change; Data points; data-driven model; Differential equations; Emissions; Farm buildings; Farms; Fluctuations; Greenhouse effect; Learning algorithms; Machine learning; Neural networks; Nitrogen; Nitrous oxide; Ozone depletion; Pollutants; Precipitation; soil ammonia and nitrate; Soil management; soil nitrogen; Soils
• ISSN: 2296-665X; 2296-665X
• DOI: 10.3389/fenvs.2024.1353049

Nitrous oxide (N2O) is one of the most significant contributors to greenhouse forcing and is the biggest contributor to ozone depletion in the 21st century, and roughly 70% of anthropogenic nitrous oxide emissions are from agriculture and soil management. Agricultural nitrous oxide emissions are shown to spike during hotspot events, and according to the data used in this study, over 78% of nitrous oxide flux occurred during just 15% of the recorded data points. Due to the complex biogeochemical processes governing nitrous oxide formation, machine learning and process-based models often fail to predict agricultural nitrous oxide flux. A novel informed neural network was developed that combined the trainability of neural networks with the rigorous differential equation-based framework of process-based models. Differential equations that explained the variability of various nitrogen-containing compounds in soil were derived, and integrated into the network loss. The informed model explained ∼85% of variation in the data and had an F1 score of 0.75, a marked improvement over the classical model explaining ∼30% of variation and having a score of 0.53. The informed network was also able to perform exceptionally well with only small subsets of the training data, having an F1 score of 0.41 with only 25% of training data. The model not only shows great promise in the remarkably accurate prediction of these hotspots but also serves as a potential new paradigm for physics-informed machine learning techniques in environmental and agricultural sciences.