Scholarlink

<p>Urban forests deliver critical ecosystem services, particularly in tropical cities where ecological variability and urban stressors interact. However, the extent to which specific tree traits influence service delivery remains underexplored. This study applied an integrated sensitivity framework, combining Morris One-at-a-Time (MOAT), Variance Decomposition (VD), and Bin-wise Standardized Linear Regression (BSLR) to evaluate global and conditional drivers of carbon, pollution, and hydrological services simulated by the i-Tree Eco model in Rizal Park, Manila, Philippines. We utilized Latin Hypercube Sampling (LHS) to generate individual synthetic tree records, preserving empirical trait correlations to ensure biological plausibility. Results showed diameter at breast height (DBH) was the dominant driver of carbon storage (MOAT µ* = 496.10; VD S₁ = 0.947). For carbon sequestration (CSeq), sensitivities were more distributed across performance tiers, reflecting a complex interplay between tree size and canopy-level traits. For air pollution removal (PM<sub>10</sub> and PM<sub>2.5</sub>), Leaf Area Index (LAI) emerged as the primary global driver of output variance (S<sub>1</sub> ≈0.57–0.58), whereas crown structural variables like crown missing (CM) were identified as critical inhibitors in low-performance scenarios Hydrological regulation was similarly governed by functional leaf area (S<sub>1</sub> ≈0.48–0.52), with LAI exhibiting massive, standardized coefficients in high-performance tiers (e.g., β<sub>PET</sub> = 22,157.09, <i>p</i> &lt; 0.001). BSLR analysis revealed a bifurcated two-stage sensitivity structure: structural integrity (low CM) was the primary prerequisite for baseline services, whereas LAI act as the performance ceiling for high-magnitude delivery. Explanatory power for these relationships was high, with Adjusted R<sup>2</sup> values ranging from 0.89 for pollution removal to near-perfect linearity (≈ 1.00) for hydrological services. These findings underscore the necessity of integrating structural and functional traits to optimize nature-based solutions and refine trait-based monitoring in tropical urban forestry.</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

Trait-based sensitivity analysis of urban trees using the i-tree eco model: insights from a tropical urban green space

  • 期刊论文

  • 发表日期：

    2026年5月12日

查看全文
  • Wencelito Palis Hintural,
  • Byung Bae Park

摘要

Urban forests deliver critical ecosystem services, particularly in tropical cities where ecological variability and urban stressors interact. However, the extent to which specific tree traits influence service delivery remains underexplored. This study applied an integrated sensitivity framework, combining Morris One-at-a-Time (MOAT), Variance Decomposition (VD), and Bin-wise Standardized Linear Regression (BSLR) to evaluate global and conditional drivers of carbon, pollution, and hydrological services simulated by the i-Tree Eco model in Rizal Park, Manila, Philippines. We utilized Latin Hypercube Sampling (LHS) to generate individual synthetic tree records, preserving empirical trait correlations to ensure biological plausibility. Results showed diameter at breast height (DBH) was the dominant driver of carbon storage (MOAT µ* = 496.10; VD S₁ = 0.947). For carbon sequestration (CSeq), sensitivities were more distributed across performance tiers, reflecting a complex interplay between tree size and canopy-level traits. For air pollution removal (PM10 and PM2.5), Leaf Area Index (LAI) emerged as the primary global driver of output variance (S1 ≈0.57–0.58), whereas crown structural variables like crown missing (CM) were identified as critical inhibitors in low-performance scenarios Hydrological regulation was similarly governed by functional leaf area (S1 ≈0.48–0.52), with LAI exhibiting massive, standardized coefficients in high-performance tiers (e.g., βPET = 22,157.09, p < 0.001). BSLR analysis revealed a bifurcated two-stage sensitivity structure: structural integrity (low CM) was the primary prerequisite for baseline services, whereas LAI act as the performance ceiling for high-magnitude delivery. Explanatory power for these relationships was high, with Adjusted R2 values ranging from 0.89 for pollution removal to near-perfect linearity (≈ 1.00) for hydrological services. These findings underscore the necessity of integrating structural and functional traits to optimize nature-based solutions and refine trait-based monitoring in tropical urban forestry.