Rice is a major staple food for almost 50% of the human population in the world. Paddy rice fields, where rice plants are cultivated in flooded/inundated soils, are widely distributed across the globe, ranging from single paddy rice crop in a year (single cropping system) in temperate zone to triple paddy rice rice crops in a year in the moist tropic regions (triple cropping system). The information on the area, spatial distribution and dynamics of paddy rice agriculture is important for assessment of food security, management of water resources, and estimation of greenhouse gas (methane) emissions as well as transmission of zoonotic infectious diseases. Satellite remote sensing is one of important approaches to measure and monitor paddy rice fields across local to global scales. Over the past few decades, most research projects have mapped croplands as one of many land cover types through analyses of AVHRR, SPOT-VEGETATION, MODIS, MERIS, and Landsat data. Through the support of NASA LCLUC program, our research group has investigated the spectral characteristics of paddy rice fields over the rice plant growing seasons along the latitudinal gradient, and has assessed the potential of time series images from coarse and moderate resolution sensors such as SPOT-4 VEGETATION and MODIS sensors to identify and map paddy rice fields. Annual regional maps of paddy rice fields in the monsoon Asia have been generated from analyses of MODIS images at 500-m spatial resolutions (Figure 1), based on the phenology- and pixel-based paddy rice mapping algorithms (PPPM), here we called the RICE-MODIS mapping tool.

Figure 1. Spatial distribution of paddy rice fields in monsoon Asia in 2010 at 500-m spatial resolution, as derived from the RICE-MODIS mapping tool.

Figure 2. Spatial distribution of paddy rice fields in northeastern Asia (Japan, North Korea, South Korea, and Northeast China) in 2014 at 30-m spatial resolution, as derived from Landsat 8 OLI images in 2014.

It is important to note that many paddy rice fields are much small in field sizes, because of complex topography constrain and small household farmers, thus, there is a need to identify and map paddy rice fields using fine resolution images such as Landsat. To date, we have evaluated the time series images from Landsat 5 and 7 (TM/ETM+) to identify and map paddy rice fields at the selected single- and double- cropping areas, and the resultant RICE-Landsat mapping tool has been documented. Based on this approach, we generated a 30-m paddy rice map in NE Asia (Figure 2), including northeastern (NE) China, South Korea, North Korea, and Japan, by using all the time series Landsat 8 OLI imagery in 2014, the phenology- and pixel-based paddy rice mapping (PPPM) algorithm, and the cloud-computing technology with the GEE platform. We found that NE China, Japan, South Korea and North Korea accounted for 60%, 20%, 14 % and 6 % of the paddy rice area in the region. The resultant paddy rice map in 2014 at 30-m spatial resolution (Figure 2) provides a detailed map for the studies of food security, water resource management, and climate in these four countries and regions. This study clearly layouts the foundation for us to work towards global-scale mapping of paddy rice fields at 30-m spatial resolution in the near future. For more details about this study please check our recent publication in Remote Sensing of Environment (Dong et al. 2016). Read more about the project