Modelling aboveground biomass of oil palm using destructive method and remote sensing data

Abstract

Biomass serves as an important indicator to assess the role of oil palm in the global carbon cycle, particularly its contribution towards carbon sequestration. Indonesia is a country that has the largest palm oil plantation, and is the second largest country to export CPO (Crude Palm Oil) after Malaysia. As the world market demand for palm oil increases, Indonesia is developing large oil palm plantations. However, information about biomass or carbon stocks contained in oil palm trees is still limited. The study quantified the aboveground biomass (AGB) of oil palm trees in South Sulawesi, Indonesia using harvesting method, allometric equations and remote sensing techniques. Nine oil palm trees ranging from three trees o f young (1- 3 years), three trees of intermediate (4-10 years) and three trees of matured (11-20 years) trees were harvested, their wet and dry biomass for different components from the stems, fronds, leaflets, fruit bunches as well as flowers were obtained. In addition, 96 trees were also sampled to get Diameters o f Breast Height (DBH), height and age information. All the information were used to develop specific allometric equations to estimate dry aboveground biomass of young, intermediate and matured oil palm trees. The use o f allometric models resulted in high accuracy when AGB estimated from the equations was compared with DBH and height. Since harvesting method and allometric equations can only be used to get AGB at local (plot) scale, remote sensing data o f Advanced Land Observing Satellite Phase Array type L-band Synthetic Aperture Radar/ALOS PALSAR were used to up-scale AGB to the entire study area. Dry AGB obtained from the harvesting method was 0.75 t ha'1, 22.17 t ha’1 and 105.41 t ha’1 for young, intermediate and matured trees respectively. The allometric equations with dbh parameter produced 0.71 t h a '1, 20.15 t ha'1, 107.41 t h a '1, and dbh with height parameters have produced 1.40 t h a'1, 27.20 t ha'1, 248.52 t ha' 1 for young, intermediate and matured trees respectively. Manipulation of HH polarization, (HH + HV)/2 and ^/(HHxHV) produced better correlation with AGB (R2 between 0.53 to 0.61). Empirical models developed with these manipulation polarizations were used to estimate the AGB in South Sulawesi. Total AGBs of the area for intermediate trees ranged between 29.94 t ha’1 to 31.51 t ha’1 whereas it was between 68.32 t ha' 1 to 71.29 t ha' 1 for matured oil palm trees. AGB estimate from ALOS PALSAR showed a 24.5 to 28 percent difference in comparison to AGB obtained via allometric equations for intermediate and matured palms. The results (AGB) obtained in this study have a potential to inform decision makers to impose better land management in oil palm plantation so to alleviate climate change.