Maxent Modeling of the Habitat Distribution of the Critically Endangered Pterocarpus indicus Willd. forma indicus In Mindanao, Philippines
Journal of Biodiversity and Environmental Sciences (JBES)
Posted: 7 May 2020
Date Written: March 23, 2017
Abstract
A current and projected suitable habitat distribution models for Pterocarpus indicus Willd. forma indicus were generated using Maximum Entropy Modeling algorithm (MaxEnt). The Receiver Operating Characteristic (ROC) - Area Under Curve (AUC) of the training and test data were 0.854 and 0.920 respectively. It was highly above the random prediction AUC of 0.5; therefore the model performance was good, reasonable and valid. The predicted suitable habitat distribution of Smooth Narra was heavily influenced by climatic variables. The variable with largest contribution was Mean Temperature of Warmest Quarter (MTWQ) with 31.2%. It was followed by Soil with 20.3%. Annual Precipitation (AP) and Precipitation of Driest Quarter (PDQ) belonged to 3rd and 5th in contribution rank with 12.8% and 8.8% respectively. This study also found out that the spatial pattern of distribution of suitable habitats is clustered. The study also predicted changes with suitable area coverage in terms of land class, protected areas and administrative boundaries would likely occur as climatic conditions change.
Keywords: MaxEnt, Habitat suitability, Species distribution modeling, Niche modeling, Smooth Narra, Protected areas
Suggested Citation: Suggested Citation