Agrochemicals, GHG Emissions, and GDP in Southeast Asia: A Machine Learning Approach with Hierarchical Clustering

Agrochemical use, GHG emissions, and gross domestic product (GDP) vary widely across Southeast Asia, making the region suitable for cluster-based sustainability analysis. This study applies hierarchical clustering analysis (HCA) to classify nine Southeast Asian countries using four standardized indicators: pesticide use, nitrogen fertilizer use, GHG emissions, and GDP. Exploratory data analysis reveals significant disparities, with Brunei and Indonesia emerging as outliers due to exceptionally high input intensity and emissions, respectively. HCA identifies four distinct clusters: (1) low-input, low-emission economies (Cambodia, Laos, Myanmar); (2) moderately intensive systems (Malaysia, Thailand, the Philippines, Vietnam); (3) a high-pesticide profile (Brunei); and (4) a high-emission, high-output outlier (Indonesia). Principal Component Analysis confirms the cluster structure and highlights variation in emission efficiency. The findings show that similar agroecological contexts can yield divergent environmental outcomes, emphasizing the role of policy and technology. This study provides the first region-wide, data-driven typology of agricultural sustainability in Southeast Asia using HCA.

Data and Resources

Additional Info

Field Value
Type of ALiSEA product None
Title Agrochemicals, GHG Emissions, and GDP in Southeast Asia: A Machine Learning Approach with Hierarchical Clustering
Description Agrochemical use, GHG emissions, and gross domestic product (GDP) vary widely across Southeast Asia, making the region suitable for cluster-based sustainability analysis. This study applies hierarchical clustering analysis (HCA) to classify nine Southeast Asian countries using four standardized indicators: pesticide use, nitrogen fertilizer use, GHG emissions, and GDP. Exploratory data analysis reveals significant disparities, with Brunei and Indonesia emerging as outliers due to exceptionally high input intensity and emissions, respectively. HCA identifies four distinct clusters: (1) low-input, low-emission economies (Cambodia, Laos, Myanmar); (2) moderately intensive systems (Malaysia, Thailand, the Philippines, Vietnam); (3) a high-pesticide profile (Brunei); and (4) a high-emission, high-output outlier (Indonesia). Principal Component Analysis confirms the cluster structure and highlights variation in emission efficiency. The findings show that similar agroecological contexts can yield divergent environmental outcomes, emphasizing the role of policy and technology. This study provides the first region-wide, data-driven typology of agricultural sustainability in Southeast Asia using HCA.
Agroecology Category
  • Input reduction and recycling
  • Climate
Agroecology Keyword
  • Chemical input control
  • Carbon neutrality
  • Climate change
  • Greenhouse gas reduction
Contributing organisations Universitas Syiah Kuala, Konstanta Utama and The Bartlett School of Environment, Energy and Resources, University College London and Universitas Abulyatama
Author Fazli, Q. S., Idroes, G. M., Hilal, I. S., Hafizah, I., Hardi, I., & Noviandy, T. R.
Year 2025
Type of document Scientific & Research
Language English
Country Southeast Asia
Administrative Level 1
Administrative Level 2
Web Link https://doi.org/10.61975/gjbes.v2i2.93