Scholarlink

<p>Understanding tree mortality is crucial to understand forest dynamics and is essential for growth models and simulators. Although factors such as competition, drought, and pathogens drive mortality, their underlying mechanisms remain difficult to model. While substantial attention has focused on selecting appropriate algorithms and covariates, evaluating individual tree mortality models also requires careful selection of performance criteria. This study compares seven different metrics to assess their impact on model evaluation and selection. Results show that candidate models exhibited varying performances across metrics and that the choice of metric significantly influences the selection of the best model. When no confusion matrix was available, the area under the precision-recall curve (AUCPR) emerged as a more reliable alternative to the area under the ROC curve (AUC), offering a more informative assessment for imbalanced datasets. When a confusion matrix was available, Cohen’s Kappa coefficient (K) and Matthews correlation coefficient (MCC) outperformed accuracy-based metrics, providing a fairer evaluation of both live and dead tree classifications. These findings emphasize the importance of choosing appropriate evaluation standards to enhance mortality model assessment and ensure reliable predictions in forestry applications.</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

How performance metric choice influences individual tree mortality model selection

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月31日

查看全文
  • Aitor Vázquez-Veloso,
  • Andrés Bravo-Núñez,
  • Astor Toraño-Caicoya,
  • Hans Pretzsch,
  • Felipe Bravo

摘要

Understanding tree mortality is crucial to understand forest dynamics and is essential for growth models and simulators. Although factors such as competition, drought, and pathogens drive mortality, their underlying mechanisms remain difficult to model. While substantial attention has focused on selecting appropriate algorithms and covariates, evaluating individual tree mortality models also requires careful selection of performance criteria. This study compares seven different metrics to assess their impact on model evaluation and selection. Results show that candidate models exhibited varying performances across metrics and that the choice of metric significantly influences the selection of the best model. When no confusion matrix was available, the area under the precision-recall curve (AUCPR) emerged as a more reliable alternative to the area under the ROC curve (AUC), offering a more informative assessment for imbalanced datasets. When a confusion matrix was available, Cohen’s Kappa coefficient (K) and Matthews correlation coefficient (MCC) outperformed accuracy-based metrics, providing a fairer evaluation of both live and dead tree classifications. These findings emphasize the importance of choosing appropriate evaluation standards to enhance mortality model assessment and ensure reliable predictions in forestry applications.