MAPPING SPATIAL DISTRIBUTION OF MAIN SEASON RICE FIELDS IN EASTERN NEPAL USING MULTI-TEMPORAL LANDSAT 8 IMAGES
Samita Thapa, Janma Jaya Gairhe
Mapping rice area can be beneficial for change detection, irrigation management, climate change impact and vegetation protection and restoration programs. Remote sensing has provided a vantage means of mapping rice area. The unique physical characteristics of rice plants is that it is grown in flooded soil, which significantly affect the spectral reflectance from the rice fields. After a period of two months, the dense rice canopy cover replaces the flooded soil. This dynamic of the rice field is captured with the help of vegetation indices and are used to identify rice fields. Multi-spectral and multi-temporal data Landsat 8 data is used in the study. An algorithm that uses Normalized difference vegetation index (NDVI) and Land-surface Water Index (LSWI) derived from Landsat 16-days 30-meter data was used to differentiate paddy field from other areas. It works on the basis of sensitivity of LSWI for surface moisture and NDVI for vegetation content. This algorithm was used to detect rice fields in twelve local levels in Sunsari and Morang districts of Nepal. The results were validated using 0.44m resolution digital globe satellite imagery with 79 well-distributed control points. The overall accuracy of the method was found to be 79.746%.
Paddy, NDVI, Food Security, Remote Sensing, LSWI, GIS.