Scholarlink

<p>Bangladesh’s agricultural sector—vital to national food security and supporting nearly half the population—has seen its GDP share decline from 52% in 1972 to 11.61% in 2022, largely due to intensifying climate stressors. Although prior research has focused on crop yields, this study is the first to apply machine learning (linear regression, support vector regression, and random forest) to assess sector-specific climate vulnerabilities and renewable energy trade-offs. Drawing on data from the World Bank, NASA, and FAO (1995–2020), the analysis indicates that rising humidity strongly correlates with declining fishery output (R² = 0.991), while heat stress correlates with livestock viability, and erratic precipitation correlates with agricultural productivity (R² = 0.974). Furthermore, while renewable energy (e.g., solar irrigation) enhances resilience, it also competes for land, reducing short-term output. Collectively, these findings highlight distinct climate sensitivities across sub-sectors and inform a scalable policy framework incorporating heat-tolerant livestock, precision irrigation, and phased RNEC deployment. The proposed model offers transferable insights for other climate-vulnerable deltaic regions, where sustainability and productivity must be jointly pursued.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Bangladesh’s Climate-Stressed Agriculture: Sectoral Vulnerabilities Via Machine Learning

  • 期刊论文

  • 发表日期：

    2025年9月30日

查看全文
  • Md Ali Emam,
  • Zhang Zhixin,
  • Ehsan Elahi,
  • Saleh Shahriar,
  • Md Zakaria,
  • Jakaria Rayhan

摘要

Bangladesh’s agricultural sector—vital to national food security and supporting nearly half the population—has seen its GDP share decline from 52% in 1972 to 11.61% in 2022, largely due to intensifying climate stressors. Although prior research has focused on crop yields, this study is the first to apply machine learning (linear regression, support vector regression, and random forest) to assess sector-specific climate vulnerabilities and renewable energy trade-offs. Drawing on data from the World Bank, NASA, and FAO (1995–2020), the analysis indicates that rising humidity strongly correlates with declining fishery output (R² = 0.991), while heat stress correlates with livestock viability, and erratic precipitation correlates with agricultural productivity (R² = 0.974). Furthermore, while renewable energy (e.g., solar irrigation) enhances resilience, it also competes for land, reducing short-term output. Collectively, these findings highlight distinct climate sensitivities across sub-sectors and inform a scalable policy framework incorporating heat-tolerant livestock, precision irrigation, and phased RNEC deployment. The proposed model offers transferable insights for other climate-vulnerable deltaic regions, where sustainability and productivity must be jointly pursued.

Graphical Abstract