The Use of Landsat TM Imagery for the Application of Rubber Tree Area and Stand Volume Predictive Models in Rubber Plantations in Selangor, Malaysia

Abstract

AbstractThe requirement for information of the spatial distribution of forest resources has increased rapidly in recent years. Integration of remote sensing data with Geographical Information System (GIS) hold the keys to contribute to effective mapping and monitoring of forest resources. Globally, there have been tremendous efforts of mapping of forest resources, agroforestry, forest plantations, and agricultural tree crops at regional and local scales. Rubber tree (Hevea brasiliensis) is one of the important agricultural tree crops which plays roles as the main sources of natural rubber and wood products not only in Malaysia but also in other countries such as Indonesia, Thailand, India, Vietnam, and China. Effective management and appropriate policy for this tree crop require reliable information on resource dynamics and forecasts of resource availability. The need for inventories and effective monitoring prompted this study into utilising Landsat Thematic Mapper (Landsat TM) for the application of rubber tree area and stand volume predictive models which was previously developed by Suratman et al. (Int For Rev 6(1):1–13, 2004). In this study, the application of predictive models has produced thematic maps of spatial distribution of rubber tree areas and stand volume for a case study area of rubber tree plantations in Selangor, Malaysia. The work in this study has demonstrated the development of methods for mapping rubber plantation by linking ground information with model projections based on satellite data. Additionally, the maps produced in this work demonstrates a spatial distribution and availability of rubber tree resources across landscapes. This information is needed for effective resource planning aimed at maximizing the potential benefits of rubber tree crops for wood and natural rubber supply in Malaysia.KeywordsRubber treeLandsat TMPredictive modelsModel applicationRegression analysis