Rice is the second most harvested staple food in the world and the leading staple food in the Asian region. Rice can be contributed to food problems as well as poverty alleviation because millions of small farmers grow millions of hectares of rice in the Asian region and there are landless workers who generate some income by working on these farms. 60% of the global population and 90% of the world rice production is derived from the Asian continent (Geert Claessens).
Rice monitoring and mapping is very important for food security, environmental sustainability, water security, greenhouse gas emission and also economically. Most of the countries in the Asian region use statistical survey method to collect rice paddy data from community level to national level. These statistical data sources have some limitations to meet up the needs of science and policy researchers. They need geospatial databases of rice agriculture with updated spatial and temporal resolution (Xiao, et al., 2006).
Remote Sensing (RS) is becoming an essential tool to monitor, map and observe rice growing over large areas, at repeated time intervals (Son, N.T., et al., 2012). According to review article of Remote Sensing of Rice Crop Areas by Kuenzer, C., & Knauer, K., 2013 Remote Sensing combined with Geographical Information System (GIS), can provide reliable information for a variety of purposes related to rice farming as follows.
Mapping and monitoring the extent of rice growing ecosystems
Monitoring and assessment of rice growth and health status
Assessment of cropping pattern and cropping system efficiency
Estimation of crop-growth related parameters
Input and improvement and extension of crop growth and yield models etc,
High temporal and medium spatial resolution optical remote sensing vegetation indices Normalized Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) and Land Surface Water Index (LSWI) are utilized to map and monitor rice agriculture in small scale in global, continental and sub continental level (Sakamoto, T., et al., 2005; Xia, et al., 2006). Moderate Resolution Imaging Spectroradiometer (MODIS) sensor data was the major data source for those studies and various time series image analysis techniques were applied to derive rice crop acreage and crop phenology such as wavelet analysis, Artificial Neural Network (ANN), Time Series Regression and Time Series Image Classification etc., (Kuenzer, C., & Knauer, K., 2013; Jonai, H., and Takeuchi, W., 2013; Sakamoto, T., et al., 2005; Son, N.T., et al., 2012; Xia, et al., 2006).
MODIS Gridded Vegetation Product
MODIS land science team had developed Global MODIS vegetation indices; which were designed to provide consistent spatial and temporal comparisons of vegetation conditions. Visible red (620 – 670 nm), near infrared (NIR) (841 – 876 nm) and visible blue (459 – 479 nm) surface reflectance of MODIS sensor data are used to make global NDVI...