Title: Role of deep learning for prediction of greenhouse gas emission from agriculture: enabling technology
Authors: Pranali K. Kosamkar; Vrushali Y. Kulkarni
Addresses: School of Computer Engineering and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India ' School of Computer Engineering and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India
Abstract: Trends in population growth and food demand need to be ensured to produce an adequate amount of food for an increasing population. So, different practices are taken by farmers without considering their adverse effect on the environment to meet the required needs. Such practices often lead to increased greenhouse gases (GHGs) emission. Agriculture sector contributes to GHGs emission through livestock, soil, use of fertiliser, etc. This study presents a systematic work done in the GHGs from the agriculture sector. The study shows that there is a correlation between agriculture activities like land use, application of nitrogen fertiliser, etc. and GHGs emission from agriculture. The study revealed that the GHGs emission can be reduced by adapting the different management practices and also by adapting technologies like artificial intelligence and deep learning. We also presented the proposed architecture using long short-term memory network (LSTM) deep learning model for analysis and prediction of GHGs emission from agriculture. We have prepared different approaches for measurement, analysis, and prediction of GHGs emission. Based on the study we presented key finding related to GHGs emission.
Keywords: deep learning; agriculture; greenhouse gas emission; climate change.
International Journal of Agriculture Innovation, Technology and Globalisation, 2021 Vol.2 No.2, pp.97 - 107
Received: 08 Jun 2020
Accepted: 07 Jul 2020
Published online: 16 Dec 2021 *