Mapping Rubber Tree Stand Age Using Pléiades Satellite Imagery: A Case Study in Talang District, Phuket, Thailand
The rubber stand age information is an important variable for determining the distribution of carbon pools and fluxes in rubber plantation ecosystems as well as for production management. This study demonstrates the capability of high spatial resolution satellite imagery with the help of feature selection (i.e., Sequential Forward Floating Selection) to improve the accuracy of rubber stand age mapping. It was found that the Pléiades imagery with the help of Sequential Forward Floating Selection can successfully classify rubber stands age. The total testing accuracy was improved from 94.07% to 94.92% and 96.61% after applying the Principal Component and the Sequential Forward Floating Selection algorithms, respectively. Since the methodology proposed in this study can accurately classify 3 classes of rubber stand age, it is anticipated that this methodology can be used as a guideline for rubber tree stand age mapping in other study areas.