Scholarlink

<p>As a world-class urban agglomeration, the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) has experienced substantial land-cover restructuring driven by urbanization, necessitating high temporal resolution monitoring to capture dynamic surface processes. However, traditional land-cover products cannot capture intra-annual dynamics effectively due to their limited update cycles. Therefore, this study generates a publicly accessible dynamic land-cover dataset for the GBA spanning 2000–2022, with a 30-meter spatial resolution and a 15-day temporal resolution. This dataset achieves an overall accuracy of 97.46%, an average accuracy of 94.5%, an F1 score of 95.19%, and a kappa coefficient of 96.78%. The validation results indicated that this dataset significantly enhances the detection of short-term land cover transitions and facilitates the characterization of critical environmental patterns. We anticipate it will serve as a baseline for monitoring agricultural growing cycles, tracking vegetation phenological shifts, and identifying disturbances in forests and croplands caused by natural hazards or anthropogenic activities.</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

Dynamic land-cover maps in the Guangdong-Hong Kong-Macao Greater Bay Area at 30 m spatial resolution, 2000-2022

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月18日

查看全文
  • Yue Li,
  • Qiang Liu,
  • Shuang Chen,
  • Lingfeng Lv,
  • Xinpei Han,
  • Run Jia,
  • Lingchen Bu,
  • Xiaotong Zhang

摘要

As a world-class urban agglomeration, the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) has experienced substantial land-cover restructuring driven by urbanization, necessitating high temporal resolution monitoring to capture dynamic surface processes. However, traditional land-cover products cannot capture intra-annual dynamics effectively due to their limited update cycles. Therefore, this study generates a publicly accessible dynamic land-cover dataset for the GBA spanning 2000–2022, with a 30-meter spatial resolution and a 15-day temporal resolution. This dataset achieves an overall accuracy of 97.46%, an average accuracy of 94.5%, an F1 score of 95.19%, and a kappa coefficient of 96.78%. The validation results indicated that this dataset significantly enhances the detection of short-term land cover transitions and facilitates the characterization of critical environmental patterns. We anticipate it will serve as a baseline for monitoring agricultural growing cycles, tracking vegetation phenological shifts, and identifying disturbances in forests and croplands caused by natural hazards or anthropogenic activities.