AI to the Rescue: New Model Farms Out Solutions to Cut Agricultural Emissions

A team led by HKUST developed an AI model to reduce agricultural NH3 emissions by up to 38%, offering strategies for sustainable agriculture aligned with UN Sustainable Development Goals

In a notable advancement, an international team led by the Hong Kong University of Science and Technology (HKUST) has developed an artificial intelligence (AI) model aimed at reducing global ammonia (NH3) emissions from agriculture. The study “Fertilizer management for global ammonia emission reduction” published in the journal Nature, presents a nuanced view of NH3 emissions, suggesting that they are lower than previously estimated.

The study also proposes methods to optimise fertiliser management, potentially reducing emissions by up to 38% without reducing the overall use of nitrogen fertilisers. This research aligns with the United Nations’ Sustainable Development Goals, focusing on poverty eradication, food security, and sustainable agriculture.

NH3 emissions, stemming mainly from agricultural activities, contribute to air and water pollution and can indirectly affect climate change by forming compounds like nitrous oxide in the soil and atmosphere. Addressing these emissions is crucial, especially since the cultivation of rice, wheat, and maize significantly contributes to the issue. The study underscores the importance of accurate, global-scale information for countries to develop tailored emission reduction strategies.

Led by Prof. Jimmy Fung Chi-Hung and Prof. Zheng Yi, the research team compiled a comprehensive dataset based on NH3 emission rates observed from 1985 to 2022. Utilising this dataset, they trained an AI model to estimate NH3 emissions, considering various geographical factors such as climate, soil characteristics, and farming practices. This model can also generate specific fertiliser management plans for different regions, offering practical solutions to reduce emissions.

Prof. Jimmy Fung Chi-Hung (second from the left) and his team showing off the new AI model

For instance, the study highlights the potential in Asia for using enhanced-efficiency fertilisers to lower NH3 emissions, especially in wheat cultivation, where temperature plays a significant role. The model’s findings suggest that optimising fertiliser use, including timing and nutrient blend adjustments, and adopting appropriate planting and tillage practices, could reduce NH3 emissions from key crops by up to 38%.

However, the study also projects a potential increase in global NH3 emissions from cropland of 4.0% to 5.5% over the next 30 years until 2060. While the proposed emission reductions are promising, they also underscore the need for sustained and concerted efforts to address this challenge effectively.

Prof. Jimmy Fung highlighted the practical implications of the study, particularly the challenges in emission reduction efforts, such as high costs and the small scale of farms. He emphasised the potential of the detailed global NH3 emission data provided by the study to inform policy and management practices, aiming to reduce pollution and enhance food security.

This research may offer a significant contribution to the understanding of NH3 emissions and presents viable strategies for their reduction. However, it also acknowledges the complexities and long-term challenges in managing global emissions, emphasising the need for informed and strategic approaches in policy and practice.

You can read the full paper in Nature.

Staff Writer

Our in-house science writing team has prepared this content specifically for Lab Horizons

Leave a Reply

Your email address will not be published. Required fields are marked *