Scholarlink

<p>Plastic film mulching has been widely adopted to enhance crop yields in arid and semi-arid regions, but it has also resulted in severe environmental pollution and altered water and heat cycles in farmland ecosystems. However, the scarcity of training samples hinders large-scale mapping of plastic-mulched farmland (PMF) distributions. Here we generated the first 10-m PMF maps for the Chinese Loess Plateau spanning the period 2019–2021 (PMF-LP) by coupling the automatic training sample generation and classifier transfer methods. The resultant maps were validated using independent samples and showed satisfactory accuracies with F1-scores ranging from 0.80 to 0.86. The estimated PMF areas derived from the PMF-LP demonstrated good agreement with agricultural census data at the municipal level (R² ≥ 0.87). This is the first attempt to map PMF distributions at 10-m resolution on the Loess Plateau. The datasets generated in this study will be valuable for pollution assessments, yield forecasting, and greenhouse gas estimations.</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

A 10-m resolution dataset of plastic-mulched farmland distributions on the Chinese Loess Plateau (2019–2021)

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月11日

查看全文
  • Cheng Zhao,
  • Yadong Luo,
  • Xiangyu Chen,
  • Linxian Jiang,
  • Zhao Wang,
  • Hao Feng,
  • Qiang Yu,
  • Jianqiang He

摘要

Plastic film mulching has been widely adopted to enhance crop yields in arid and semi-arid regions, but it has also resulted in severe environmental pollution and altered water and heat cycles in farmland ecosystems. However, the scarcity of training samples hinders large-scale mapping of plastic-mulched farmland (PMF) distributions. Here we generated the first 10-m PMF maps for the Chinese Loess Plateau spanning the period 2019–2021 (PMF-LP) by coupling the automatic training sample generation and classifier transfer methods. The resultant maps were validated using independent samples and showed satisfactory accuracies with F1-scores ranging from 0.80 to 0.86. The estimated PMF areas derived from the PMF-LP demonstrated good agreement with agricultural census data at the municipal level (R² ≥ 0.87). This is the first attempt to map PMF distributions at 10-m resolution on the Loess Plateau. The datasets generated in this study will be valuable for pollution assessments, yield forecasting, and greenhouse gas estimations.