Assessing the Performance of the AUC Index for Dieback Sensitivity Mapping in the Forests of Lordegan

Forest dieback is one of the most concerning phenomena affecting forest ecosystems in Iran, particularly in the Zagros region. This study aimed to evaluate the accuracy of dieback sensitivity maps in the forests of Lordegan County using two machine learning models: Support Vector Machine (SVM) and Multilayer Perceptron Artificial Neural Network (MLP). The Area Under the Curve (AUC) index was employed as the performance evaluation criterion for both training and validation phases. The AUC results indicated that both SVM and MLP models performed well in identifying dieback-susceptible areas. These findings can contribute to improved forest resource management and proactive planning in response to forest dieback.

کلیدواژه ها:

Dieback ، AUC Index ، Zagros Forest ، Support Vector Machine (SVM)