GHG Global Emission Prediction of Synthetic N Fertilizers Using Expectile Regression Techniques

Agriculture accounts for a large percentage of nitrous oxide (N2O) emissions, mainly due to the misapplication nitrogen-based fertilizers, leading an increase in greenhouse gas (GHG) footprint. These emissions are direct nature, released straight into atmosphere through nitrification and denitrification, or indirect nitrate leaching, runoff, N2O volatilization processes. largely ascribed agricultural sector, which represents threat sustainability food production, subsequent radical contribution climate change. In this connection, it is crucial unveil relationship between synthetic N fertilizer global use emissions. To end, we worked on dataset drawn from recent study, estimates according each country, by Intergovernmental Panel Climate Change (IPCC) guidelines. Machine learning tools considered great explainable techniques when dealing with air quality problems. Hence, our work focuses expectile regression (ER) based-approaches predict based use. contrast classical linear (LR), method allows heteroscedasticity omits parametric specification underlying distribution. ER provides complete picture target variable’s distribution, especially tails interest, heavy-tailed distributions. work, applied kernel estimator (KERE) The results outline both flexibility competitiveness ER-based regard state-of-the-art approaches.