Land-cover classification in Kalimantan by polarimetric PALSAR

The need for accurate estimation of forest cover information in Kalimantan is currently increasing because of the growing recognition of the potential role of forests in helping mitigate effects of climate change and global warming. Synthetic aperture radar (SAR) technology with different polarizations has been widely used for large-scale forest monitoring and land cover mapping. This study investigates the ability of ALOS PALSAR L-band data at quad polarization (four different polarizations: HH, HV, VH, and VV) and 15m resolution to provide useful information on forest and other land cover types for the purpose of accuracy mapping land cover types. The powerful source of polarimetric information allows us to more accuracy mapping the land-cover types. Several scenes of PALSAR level 1.1 Single Look Complex (SLC) data have been acquired in 2009 and 2010 for the south Kalimantan, Indonesia. In addition, two scenes of ALOS AVNIR-2 were also acquired for ancillary data set. Two land cover classification methods: maximum likelihood classification (MLC) method and subspace method were employed for land cover mapping. Our results indicate that when combining the polarimetric coherency T3 matrix with intensity images, the classification accuracy is higher than that using only four bands (HH, HV, VH, and VV) polarization data.