Scholarlink

<p>Understanding quantitative wetland health status due to environmental change can help assess the risk of wetlands losing their ecosystem services. This study introduces the Standardized Wetland Area Index (SWAI) to quantitatively monitor the health status of Anzali wetland, using various hydroclimatic variables and state-of-the-art data-driven models, including Random Forest (RF), Multivariate Adaptive Regression Splines (MARS), and Support Vector Regression (SVR). The hydroclimatic datasets (2002–2019) were collected from in-situ and remotely sensed information. The Quantitative Wetland Risk Assessment Index (QWRAI) is also presented to assess the risk of wetland loss, calculated from the total scores of five classified criteria, including wetland area/water level, land use, SWAI, and wetland habitat. The results showed that under comparable modeling conditions, the Random Forest (RF) model yielded the highest simulation accuracy for SWAI, with KGE values of 0.87 (calibration) and 0.83 (validation), representing approximately 6% improvement over SVR and 9–10% over MARS. In addition, according to the indicators’ results, the Anzali wetland is situated in a high-risk group of wetlands. The outcomes highlight that these indicators are suitable for rapid assessment of wetland health/risk conditions on a monthly/annual basis and fill a vital gap in wetland evaluation metrics. For the first time, these indicators have been specifically developed through this data-driven methodology to align with the Sustainable Development Goals (SDGs) and the Ramsar Strategic Plan for coastal wetlands. They equip decision-makers with practical, evidence-based tools to strengthen wetland conservation and restoration policies, especially by focusing on ecological risk mitigation and coastal wetland management.</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

Proposing New Wetland Health and Risk Indicators Using Hydroclimatic Variables and Data-Driven Models for a Coastal Wetland

  • 期刊论文

  • 发表日期：

    2025年12月18日

查看全文
  • Ali Haji Elyasi,
  • Mohsen Nasseri,
  • Peyman Badiei

摘要

Understanding quantitative wetland health status due to environmental change can help assess the risk of wetlands losing their ecosystem services. This study introduces the Standardized Wetland Area Index (SWAI) to quantitatively monitor the health status of Anzali wetland, using various hydroclimatic variables and state-of-the-art data-driven models, including Random Forest (RF), Multivariate Adaptive Regression Splines (MARS), and Support Vector Regression (SVR). The hydroclimatic datasets (2002–2019) were collected from in-situ and remotely sensed information. The Quantitative Wetland Risk Assessment Index (QWRAI) is also presented to assess the risk of wetland loss, calculated from the total scores of five classified criteria, including wetland area/water level, land use, SWAI, and wetland habitat. The results showed that under comparable modeling conditions, the Random Forest (RF) model yielded the highest simulation accuracy for SWAI, with KGE values of 0.87 (calibration) and 0.83 (validation), representing approximately 6% improvement over SVR and 9–10% over MARS. In addition, according to the indicators’ results, the Anzali wetland is situated in a high-risk group of wetlands. The outcomes highlight that these indicators are suitable for rapid assessment of wetland health/risk conditions on a monthly/annual basis and fill a vital gap in wetland evaluation metrics. For the first time, these indicators have been specifically developed through this data-driven methodology to align with the Sustainable Development Goals (SDGs) and the Ramsar Strategic Plan for coastal wetlands. They equip decision-makers with practical, evidence-based tools to strengthen wetland conservation and restoration policies, especially by focusing on ecological risk mitigation and coastal wetland management.