Mapping the Spatial Distribution of Paddy Rice using Time-series TERAA-MODIS Data

In order to prepare information about spatial distribution of rice fields, Multi-Spectral and Multi-Temporal data can be helpful because in addition to the rice, all fields could be covered by mixture of water and soil regarding the time of crop calendar. In this study, we developed a special mapping algorithm that uses NDVI and LSWI2105 time series data derived from MODIS 16-day 250 meter vegetation indices (MOD13Q1) product of MODIS imagery to identify paddy rice fields. This algorithm works based on the sensitivity of LSWI2105 to the surface moisture and NDVI to the vegetation chlorophyll content. As a result, during the period of Rice Transplanting stage because of the field flooding, LSWI2105 will be increased and after the rice growing it will be decreased, and at the same time NDVI has a reverse behavior. In this research two methods has been developed to define the relationship between NDVI and LSWI2105 to detect the location of paddy rice fields in North part of Iran in 2011. Our results were validated with ground field data at 183 well-distributed sample points. The overall Accuracy of two methods are 67.21% and 86.87% respectively. The results of this study indicated that the paddy rice mapping based on MOD13Q1 could potentially be applied at moderate spatial scales to monitor paddy rice agricultural fields on a timely basis.