Scholarlink

<p>Reliable species-level monitoring of mangrove ecosystems is increasingly required for environmental assessment and regulatory decision-making, yet a methodological gap persists between remote sensing outputs and ecological risk evaluation. This study proposes an integrated assessment framework that links multi-sensor satellite imagery with ecological niche modeling to support precision monitoring and threshold-based management. Using Sentinel-2 and SPOT-6 data combined with an object-based U-Net model, we achieved high species-level classification accuracy (OA = 89.02%, Kappa = 0.82) for approximately 483.6&#xa0;ha of mangroves in Quanzhou Bay, China. Generalized Additive Models (GAMs) quantified key environmental thresholds, revealing a structured zonation scheme: <i>Kandelia obovata</i> prevails in highly productive, hydrologically connected zones (NDVI &gt; 0.43; NDTI &gt; 0.37), <i>Aegiceras corniculatum</i> persists in moderately stressed transitional habitats, while <i>Avicennia marina</i> exhibits critical niche compression, with occurrence probability sharply declining beyond approximately 200&#xa0;m from tidal creeks. These quantified thresholds provide measurable indicators for ecological monitoring, enabling the early identification of vulnerable species and habitat degradation risks. By integrating species-level distribution mapping with quantified environmental thresholds, the proposed approach operationalizes remote sensing products into a tiered set of management-relevant indicators, including mangrove habitat extent, fine-scale species distribution patterns, and key environmental drivers of species zonation. These evaluation-ready metrics support targeted hydrological restoration, species-specific rehabilitation planning, and invasive species control. This framework demonstrates strong transferability for environmental condition assessment and adaptive monitoring in coastal wetlands under increasing anthropogenic pressure.</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

Fine-scale mangrove species monitoring and ecological threshold assessment for coastal management in Quanzhou Bay, China

  • 期刊论文

  • 发表日期：

    2026年4月6日

查看全文
  • Siming Chen,
  • Fei Chen,
  • Longhui Niu,
  • Zhong Deng,
  • Kebin Wu,
  • Wenbin Li

摘要

Reliable species-level monitoring of mangrove ecosystems is increasingly required for environmental assessment and regulatory decision-making, yet a methodological gap persists between remote sensing outputs and ecological risk evaluation. This study proposes an integrated assessment framework that links multi-sensor satellite imagery with ecological niche modeling to support precision monitoring and threshold-based management. Using Sentinel-2 and SPOT-6 data combined with an object-based U-Net model, we achieved high species-level classification accuracy (OA = 89.02%, Kappa = 0.82) for approximately 483.6 ha of mangroves in Quanzhou Bay, China. Generalized Additive Models (GAMs) quantified key environmental thresholds, revealing a structured zonation scheme: Kandelia obovata prevails in highly productive, hydrologically connected zones (NDVI > 0.43; NDTI > 0.37), Aegiceras corniculatum persists in moderately stressed transitional habitats, while Avicennia marina exhibits critical niche compression, with occurrence probability sharply declining beyond approximately 200 m from tidal creeks. These quantified thresholds provide measurable indicators for ecological monitoring, enabling the early identification of vulnerable species and habitat degradation risks. By integrating species-level distribution mapping with quantified environmental thresholds, the proposed approach operationalizes remote sensing products into a tiered set of management-relevant indicators, including mangrove habitat extent, fine-scale species distribution patterns, and key environmental drivers of species zonation. These evaluation-ready metrics support targeted hydrological restoration, species-specific rehabilitation planning, and invasive species control. This framework demonstrates strong transferability for environmental condition assessment and adaptive monitoring in coastal wetlands under increasing anthropogenic pressure.