The Use of Machine Learning Algorithms to Identify Drivers of Atmospheric Ammonia concentrations
| Author | Khagendra Raj Baral | |
| Author | John McIlroy | |
| Author | Yam Kanta Gaihre | |
| Date of acession | 2024-02-13T09:45:54Z | |
| Date of availability | 2024-02-13T09:45:54Z | |
| Date of issue | 2022-12 | |
| Abstract | Rising animal product demand is driving ammonia (NH3) emissions from livestock manure, posing threats to human health, air and water quality, and biodiversity. This study investigated the relationships between weather conditions and NH3 concentrations in Northern Ireland's temperate grasslands to inform mitigation strategies. Using over 3 years of high-resolution atmospheric NH3 data and weather parameters, we employed a random forest machine learning algorithm to identify key drivers. The results showed: NH3 concentrations are intricately linked to both farm management and environmental factors. Wind speed, air temperature, wind direction, and rainfall emerged as crucial variables. Low wind speed, high temperature, and minimal rainfall conditions amplify NH3 concentrations. While including farm activities like fertilizer/manure spreading could potentially improve prediction, data limitations require cautious integration to avoid compromising the model's effectiveness. This study highlights the value of machine learning in unraveling complex NH3 drivers and paves the way for developing data-driven mitigation strategies for sustainable livestock management. | |
| URL | https://hub.ifdc.org/handle/20.500.14297/2835 | |
| Language | en | |
| Subject | Machine learning | |
| Subject | Farm management | |
| Title | The Use of Machine Learning Algorithms to Identify Drivers of Atmospheric Ammonia concentrations |