Scholarlink

<p>Accurately delineating bioclimatic zones is critical for sustainable resource management, biodiversity conservation, and addressing environmental challenges in semi-arid regions. While traditional deterministic methods for bioclimatic mapping have been widely applied, they are limited in their ability to explicitly represent uncertainty and variability, factors that are increasingly recognized as essential in environmental modeling. To address these limitations, this study integrates satellite-derived data with Bayesian Belief Networks (BBNs) to model and delineate bioclimatic zones in the semi-arid landscape of the Central Zagros Mountains, Iran. Key datasets from 2016, including the Digital Elevation Model (DEM), precipitation data from the Global Precipitation Measurement (GPM) mission, Normalized Difference Vegetation Index (NDVI), and MODIS land surface temperature (LST), were utilized, alongside the computation of the Temperature Vegetation Dryness Index (TVDI) from NDVI and LST. Sensitivity analysis revealed DEM as the most influential factor in bioclimatic zoning, outperforming other variables such as TVDI, NDVI, LST, and precipitation. The BBNs model, validated against the Pabot method, achieved robust performance metrics, with a Kappa coefficient of 0.63 and an Area Under the Curve (AUC) exceeding 0.8 for all identified zones, including arid-forest, semi-steppe, and high mountain regions. The BBNs model demonstrated versatile functionalities, including scenario analysis to forecast bioclimatic zones under varying environmental conditions and diagnostic analysis to identify the most probable scenarios leading to specific bioclimatic outcomes. This study introduces a robust probabilistic framework that integrates remote sensing data with BBNs, addressing the inherent limitations of deterministic approaches. Adaptable to diverse ecosystems, this approach provides a powerful tool for bioclimatic zone prediction, supporting informed decision-making, sustainable resource management, and biodiversity conservation efforts.</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

Integrating remote sensing and Bayesian modeling for bioclimatic mapping in the central Zagros Mountains, Iran

  • 期刊论文

  • 发表日期：

    2026年1月6日

查看全文
  • Leila Ghorbani,
  • Hossein Bashari,
  • Reza Jafari

摘要

Accurately delineating bioclimatic zones is critical for sustainable resource management, biodiversity conservation, and addressing environmental challenges in semi-arid regions. While traditional deterministic methods for bioclimatic mapping have been widely applied, they are limited in their ability to explicitly represent uncertainty and variability, factors that are increasingly recognized as essential in environmental modeling. To address these limitations, this study integrates satellite-derived data with Bayesian Belief Networks (BBNs) to model and delineate bioclimatic zones in the semi-arid landscape of the Central Zagros Mountains, Iran. Key datasets from 2016, including the Digital Elevation Model (DEM), precipitation data from the Global Precipitation Measurement (GPM) mission, Normalized Difference Vegetation Index (NDVI), and MODIS land surface temperature (LST), were utilized, alongside the computation of the Temperature Vegetation Dryness Index (TVDI) from NDVI and LST. Sensitivity analysis revealed DEM as the most influential factor in bioclimatic zoning, outperforming other variables such as TVDI, NDVI, LST, and precipitation. The BBNs model, validated against the Pabot method, achieved robust performance metrics, with a Kappa coefficient of 0.63 and an Area Under the Curve (AUC) exceeding 0.8 for all identified zones, including arid-forest, semi-steppe, and high mountain regions. The BBNs model demonstrated versatile functionalities, including scenario analysis to forecast bioclimatic zones under varying environmental conditions and diagnostic analysis to identify the most probable scenarios leading to specific bioclimatic outcomes. This study introduces a robust probabilistic framework that integrates remote sensing data with BBNs, addressing the inherent limitations of deterministic approaches. Adaptable to diverse ecosystems, this approach provides a powerful tool for bioclimatic zone prediction, supporting informed decision-making, sustainable resource management, and biodiversity conservation efforts.