Classification of rice fields in a complex land-use watershed in Northern Vietnam using RADARSAT-2 data

Due to the strong temporal backscatter signature when the rice grows above the water surface, Synthetic Aperture Radar (SAR) for paddy rice crop mapping is operational in Southern Vietnam (Mekong Delta). In Northern Vietnam, the rice mapping using SAR is a challenge and is rarely performed because of the complex land-use/land-cover. Nevertheless, the information of rice fields is important for hydrological simulation in the studied watershed. The purpose of the research is to adapt the classification methods and algorithms to monitor rice fields over a large and fragmented land-use area as the Câu River watershed, in northern Vietnam. RADARSAT-2 data were collected in dual-pol Standard and quad-pol Fine mode for two rice crop seasons. As result, RADARSAT-2 dual- and quad-pol data have demonstrated their capabilities to identify the cultivated rice fields. However, RADARSAT-2 dual-pol data in Standard mode could not identify all the rice fields identify by the quad-pol data in Fine mode.