Species distribution models: ecological explanation and prediction of an endemic and endangered plant species (Pterocarpus santalinus L.f.).
Pterocarpus santalinus L.f. (Red Sanders) is an endemic and endangered species largely confined to the southern portion of the Eastern Ghats, Andhra Pradesh, India. To understand its ecological and geographic distribution, we used ecological niche modelling (ENM) based on field sample-based distributional information, in relation to climatic and topographic datasets. Before modelling, hierarchical partitioning was used to optimize the choice of variables for better prediction and reliability. We used three ENM approaches, namely GARP, Maxent and BIOCLIM for predicting potential areas of occurrence. The ENM successfully reconstructed key features of the species geographic distribution, mainly in the forest tracts of Chittoor and Kadapa districts. GARP appeared to be more robust in prediction capabilities compared to BIOCLIM. The potential distributional area identified by these models falls mainly in regions not protected and experiencing high anthropogenic pressure owing to economic and medicinal use. The success of this model indicates that ENM-based approaches provide a promising tool for exploring various scenarios useful in the study of ecology, biogeography and conservation.
Main Authors: | , , , , |
---|---|
Format: | Journal Article biblioteca |
Language: | English |
Published: |
2012
|
Subjects: | plant ecology, biogeography, endangered species, pterocarpus santalinus, andhra pradesh, |
Online Access: | https://hdl.handle.net/10568/34512 http://www.currentscience.ac.in/Volumes/102/08/1157.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Pterocarpus santalinus L.f. (Red Sanders) is an endemic and endangered species largely confined to the southern portion of the Eastern Ghats, Andhra Pradesh, India. To understand its ecological and geographic distribution, we used ecological niche modelling (ENM) based on field sample-based distributional information, in relation to climatic and topographic datasets. Before modelling, hierarchical partitioning was used to optimize the choice of variables for better prediction and reliability. We used three ENM approaches, namely GARP, Maxent and BIOCLIM for predicting potential areas of occurrence. The ENM successfully reconstructed key features of the species geographic distribution, mainly in the forest tracts of Chittoor and Kadapa districts. GARP appeared to be more robust in prediction capabilities compared to BIOCLIM. The potential distributional area identified by these models falls mainly in regions not protected and experiencing high anthropogenic pressure owing to economic and medicinal use. The success of this model indicates that ENM-based approaches provide a promising tool for exploring various scenarios useful in the study of ecology, biogeography and conservation. |
---|