Machine learning analysis of greenhouse gas sources impacting Africa’s food security nexus

Abstract The essential need to identify the most informative sources of greenhouse gas emissions (climate change drivers) impacting the food security nexus in Africa requires a comprehensive and holistic approach. Machine learning method excels in the identification of single-variable importance, ou...

Full description

Saved in:
Bibliographic Details
Main Authors: Adusei Bofa, Temesgen Zewotir
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-14766-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849333339447623680
author Adusei Bofa
Temesgen Zewotir
author_facet Adusei Bofa
Temesgen Zewotir
author_sort Adusei Bofa
collection DOAJ
description Abstract The essential need to identify the most informative sources of greenhouse gas emissions (climate change drivers) impacting the food security nexus in Africa requires a comprehensive and holistic approach. Machine learning method excels in the identification of single-variable importance, our study complements their abilities by deriving principal components using principal component analysis (PCA) to give a significant understanding of the three primary greenhouse gases (carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O)). We used data from FAO concerning Africa from 2000 to 2021. Food household consumption emission (CH₄), burning crop residues (N₂O), and food transport (N₂O) are the key variables identified by machine learning as critical contributors to climate change drivers impacting food security. Biomass-burning emissions factor, land management emissions factor, and food supply chain emissions factor are the key principal components after subjecting the 86 to PCA. Focusing on the variables and the factors revealed by the extreme gradient, random forest, and PCA can help stakeholders develop efficient practices like promoting sustainable crop residue management and a sustainable food system that reduces post-harvest loss. Our study did not consider the potential impact of spatial effect in the identification of the key sources of greenhouse gases impacting food security, we will explore this in future works.
format Article
id doaj-art-6c0d6fb09f64483e92ae5404d66960cb
institution Kabale University
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-6c0d6fb09f64483e92ae5404d66960cb2025-08-20T03:45:53ZengNature PortfolioScientific Reports2045-23222025-08-0115111210.1038/s41598-025-14766-7Machine learning analysis of greenhouse gas sources impacting Africa’s food security nexusAdusei Bofa0Temesgen Zewotir1School of Business and Applied Sciences, Garden City University CollegeSchool of Mathematics, Statistics, and Computer Science, University of KwaZulu NatalAbstract The essential need to identify the most informative sources of greenhouse gas emissions (climate change drivers) impacting the food security nexus in Africa requires a comprehensive and holistic approach. Machine learning method excels in the identification of single-variable importance, our study complements their abilities by deriving principal components using principal component analysis (PCA) to give a significant understanding of the three primary greenhouse gases (carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O)). We used data from FAO concerning Africa from 2000 to 2021. Food household consumption emission (CH₄), burning crop residues (N₂O), and food transport (N₂O) are the key variables identified by machine learning as critical contributors to climate change drivers impacting food security. Biomass-burning emissions factor, land management emissions factor, and food supply chain emissions factor are the key principal components after subjecting the 86 to PCA. Focusing on the variables and the factors revealed by the extreme gradient, random forest, and PCA can help stakeholders develop efficient practices like promoting sustainable crop residue management and a sustainable food system that reduces post-harvest loss. Our study did not consider the potential impact of spatial effect in the identification of the key sources of greenhouse gases impacting food security, we will explore this in future works.https://doi.org/10.1038/s41598-025-14766-7AfricaExtreme gradient boostingFood securityGreenhouse gasPrincipal componentRandom forest
spellingShingle Adusei Bofa
Temesgen Zewotir
Machine learning analysis of greenhouse gas sources impacting Africa’s food security nexus
Scientific Reports
Africa
Extreme gradient boosting
Food security
Greenhouse gas
Principal component
Random forest
title Machine learning analysis of greenhouse gas sources impacting Africa’s food security nexus
title_full Machine learning analysis of greenhouse gas sources impacting Africa’s food security nexus
title_fullStr Machine learning analysis of greenhouse gas sources impacting Africa’s food security nexus
title_full_unstemmed Machine learning analysis of greenhouse gas sources impacting Africa’s food security nexus
title_short Machine learning analysis of greenhouse gas sources impacting Africa’s food security nexus
title_sort machine learning analysis of greenhouse gas sources impacting africa s food security nexus
topic Africa
Extreme gradient boosting
Food security
Greenhouse gas
Principal component
Random forest
url https://doi.org/10.1038/s41598-025-14766-7
work_keys_str_mv AT aduseibofa machinelearninganalysisofgreenhousegassourcesimpactingafricasfoodsecuritynexus
AT temesgenzewotir machinelearninganalysisofgreenhousegassourcesimpactingafricasfoodsecuritynexus