Abstract

Time series observations from the Landsat, MODIS… have provided useful datasets for identifying and monitoring agricultural processes. This study aims to develop a method monitoring the rice cropping systems in Tien Hai district, Thai Binh province using Landsat data. This study processed Landsat satellite data from January 2008 to November 2016 (45 scenes), and using the decision tree method. Classification process is performed in four main steps: (1) extraction and processing EVI, NDWI times series data; (2) mapping paddy extent; (3) mapping rice cycles; and (4) accuracy assessment. The overall accuracy of the classification was 77.3% and 80% respectively for 2010 and 2015. The results by comparisons between Landsat-derived rice area and rice area statistics have not been evaluated high. This study has demonstrated the ability to monitor the process of rice cropping by Landsat data.