Efficiency of multi-frequency, multi-polarized SAR data to monitor growth stages of oilpalm plants in Sarawak, Malaysia

Abstract

Carbon rich tropical forests of Southeast Asia have increasingly become a target for the conversion to plantation area. Plantations are economically important for the local society because they provide incentives in a short period of time. Southeast Asia has a high rate of conversion of tropical forests to various types of plantation species such as oilpalm, rubber, acacia, cashew, teak, eucalyptus etc. Therefore, monitoring of carbon loss due to the conversion of forests to various plantation species has become crucial in the global carbon cycle. Information about the age or growth stages (juvenile, young, mature and old) of plantation species is a useful parameter to calculate carbon sequestration as well as for predicting yield and a range of management practices such as determining location for thinning, harvesting and replantation. This study focuses on the comparison and efficiency of multi-frequency, multi-polarized Synthetic Aperture Radar (SAR) data to monitor various growth stages of oilpalm plants in Sarawak, Malaysia based on backscattering and polarimetric decomposition techniques. Phased Array type L-band Synthetic Aperture Radar (PALSAR) data with dual and quad polarization, Radarsat-2 data with quad polarization and TerraSAR-X (TSX) data with HH polarization have been used in this investigation. Results show that, PALSAR data with HV polarization shows highest sensitivity with the oilpalm plant's age as compared to other SAR data. TSX based texture information will also be utilized to find the important texture parameters to identify various growth stages of oilpalm plants.