Predicting future climate scenarios: a machine learning perspective on greenhouse gas emissions in agrifood systems DOI Creative Commons

Omid Behvandi,

Hamzeh Ghorbani

Frontiers in Environmental Science, Journal Year: 2024, Volume and Issue: 12

Published: Nov. 19, 2024

Global climate change is an extensive phenomenon characterized by alterations in weather patterns, temperature trends, and precipitation levels. These variations substantially impact agrifood systems, encompassing the interconnected components of farming, food production, distribution. This article analyzes 8,100 data points with 27 input features that quantify diverse aspects system’s contribution to predicted Greenhouse Gas Emissions (GHGE). The study uses two machine learning algorithms, Long-Short Term Memory (LSTM) Random Forest (RF), as well a hybrid approach (LSTM-RF). LSTM-RF model integrates strengths LSTM RF. LSTMs are adept at capturing long-term dependencies sequential through memory cells, addressing vanishing gradient problem. Meanwhile, its ensemble approach, RF improves overall performance generalization combining multiple weak learners. Additionally, provides insights into importance features, helping understand significant contributors model’s predictions. results demonstrate algorithm outperforms other algorithms (for test subset, RMSE = 2.977 R 2 0.9990). findings highlight superior accuracy compared individual being less accurate comparison. As determined Pearson correlation analysis, key variables such on-farm energy use, pesticide manufacturing, land use factors significantly influence GHGE outputs. Furthermore, this heat map visually represent coefficient between GHGE, enhancing our understanding complex interactions within system. Understanding intricate connection systems crucial for developing practices security environmental challenges.

Language: Английский

Gas content, geochemical characteristics and implications of coalbed methane from the Deep Area of Qi’Nan Coalmine in Huaibei Coalfield DOI Creative Commons
Qiang Wei, Song Chen, Wu Yi

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 16, 2024

The objective of this work is to investigate the implications geological influence factors on gas content and geochemical characteristics deep-buried (> 800 m) coalbed methane (CBM) reservoirs. Results show that bituminous coal accounts for majority, which exhibits similar maturity but differ in maceral chemical constituents. CBM reservoirs low porosity, permeability moderate temperature, with thickness 0.85–4.15 m. In addition, total 4.58–12.33 m3/t (average 8.83 m3/t). CH4 main component concentration 92.83−99.22% 96.68%), δ13CCH4 δ13DCH4 − 53.78‰–−44.62‰ 48.82‰) 223.93‰––−4.49‰ 215.37‰), respectively. All samples are mixtures thermogenic gases secondary biogenic CO2 reduction. characteristic at critical point burial depth result by positive negative effects. shows a wide range increases buried depth, while numerical values selected display complex variation characteristics. Furthermore, become heavier depth. Besides, above two parameters related Ro, max reservoir temperature.

Language: Английский

Citations

0
Predicting future climate scenarios: a machine learning perspective on greenhouse gas emissions in agrifood systems DOI Creative Commons

Omid Behvandi,

Hamzeh Ghorbani

Frontiers in Environmental Science, Journal Year: 2024, Volume and Issue: 12

Published: Nov. 19, 2024

Global climate change is an extensive phenomenon characterized by alterations in weather patterns, temperature trends, and precipitation levels. These variations substantially impact agrifood systems, encompassing the interconnected components of farming, food production, distribution. This article analyzes 8,100 data points with 27 input features that quantify diverse aspects system’s contribution to predicted Greenhouse Gas Emissions (GHGE). The study uses two machine learning algorithms, Long-Short Term Memory (LSTM) Random Forest (RF), as well a hybrid approach (LSTM-RF). LSTM-RF model integrates strengths LSTM RF. LSTMs are adept at capturing long-term dependencies sequential through memory cells, addressing vanishing gradient problem. Meanwhile, its ensemble approach, RF improves overall performance generalization combining multiple weak learners. Additionally, provides insights into importance features, helping understand significant contributors model’s predictions. results demonstrate algorithm outperforms other algorithms (for test subset, RMSE = 2.977 R 2 0.9990). findings highlight superior accuracy compared individual being less accurate comparison. As determined Pearson correlation analysis, key variables such on-farm energy use, pesticide manufacturing, land use factors significantly influence GHGE outputs. Furthermore, this heat map visually represent coefficient between GHGE, enhancing our understanding complex interactions within system. Understanding intricate connection systems crucial for developing practices security environmental challenges.

Language: Английский

Citations

0