Type: Explained

Introduction

• Researchers have harnessed machine learning to provide precise estimates of ammonia emissions stemming from rice, wheat, and maize crops.
• Their dataset allows for a crop-specific assessment of emission reduction potential, suggesting that effective fertilizer management in these crops could decrease atmospheric ammonia emissions from agriculture by up to 38%.

Ammonia Emissions in Agriculture

Ammonia (NH3) emissions primarily originate from agricultural activities, particularly livestock farming and the application of synthetic and organic fertilizers.

1. Livestock Farming: Livestock, such as cattle, poultry, and swine, produce ammonia through the breakdown of urea in their urine and faeces. Confined animal feeding operations (CAFOs) are major contributors to ammonia emissions.
2. Fertilizer Application: Ammonia is released when synthetic fertilizers containing ammonium-based compounds (e.g., ammonium nitrate) are applied to crops. Manure from livestock can also be used as organic fertilizer, contributing to ammonia emissions.

Why it matters?

• Environmental Impact: Ammonia emissions can lead to air pollution, especially in areas with intensive agriculture. It can react with other pollutants to form fine particulate matter (PM2.5) and contribute to the formation of ground-level ozone, which has adverse effects on human health and the environment.
• Acid Deposition: Ammonia can undergo atmospheric transformation and contribute to acid rain, which can harm aquatic ecosystems, forests, and infrastructure.
• Nutrient Loss: Ammonia emissions represent a loss of valuable nitrogen nutrients from agricultural systems. This can reduce the efficiency of fertilizer use and contribute to nitrogen pollution in water bodies.

Significance of Ammonia Emissions

• Environmental Impact: Atmospheric ammonia is a significant environmental pollutant, affecting ecosystems and human health globally.
• Crop-Related Emissions: A substantial portion of anthropogenic ammonia emissions, 51-60%, originates from crop cultivation. Rice, wheat, and maize are responsible for approximately half of these emissions.

Machine Learning-Based Modeling

• Researchers’ Approach: The study employed machine learning to model ammonia emissions from rice, wheat, and maize farming worldwide. This modelling considered various factors such as climate, soil characteristics, crop types, irrigation, tillage practices, and fertilization methods.
• Dataset Development: To train the model, researchers curated a dataset comprising ammonia emissions data from over 2,700 observations, gathered through a systematic review of published literature.
• Global Emission Estimate: The model’s estimates revealed that global ammonia emissions reached 4.3 teragrams (4.3 billion kilograms) in 2018.

Emission Reduction Potential

• Optimizing Fertilizer Management: By spatially optimizing fertilizer management according to the model’s guidance, ammonia emissions from the three crops could potentially be reduced by 38%.
• Strategies: The optimized strategy involves deeper placement of enhanced-efficiency fertilizers into the soil using conventional tillage practices during the growing season.

Crop-Specific Contributions

• Reduction Potential: Under the proposed fertilizer management scenario, rice crops could contribute to 47% of the total reduction potential. Maize and wheat could contribute 27% and 26%, respectively.
• Emission Projections: Without management strategies, ammonia emissions could increase by 4.6% to 15.8% by 2100, depending on future greenhouse gas emissions levels.

Conclusion

• This study showcases how machine learning can provide valuable insights into ammonia emissions from crop cultivation.
• By optimizing fertilizer management practices, substantial reductions in ammonia emissions from rice, wheat, and maize crops can be achieved, contributing to environmental sustainability.