RICE MAPPING AND PARAMETERS RETRIEVAL USING ASAR DATA IN JIANGSU PROVINCE, CHINA

摘要

This paper presents a study of multi-temporal and dual-polarization (HH, VV) ASAR data, which aims to assess the use of ASAR data for rice field mapping and monitoring. Temporal variation of backscattering coefficients of 4 categories of rice fields were derived from 2004 ASAR images of the region of Jiangsu. Ground data were collected to calculate wet biomass and LAI of rice, NOAA images were utilized to derive NDVI, and the correlation between the backscattering coefficient and those rice parameters was analyzed. Based on the backscatter behaviors of rice, a method for rice fields mapping using image ratio techniques has been developed, with accuracy higher than 80%. Four classes of rice according to varieties (hybrid or Japonica) and crop calendar have been mapped, in addition to the other main land use classes in the region. The results appear promising for the use of ASAR data on rice monitoring and rice mapping.