Paddy Monitoring in Seberang Perak, Malaysia Using Multi-Temporal Sentinel-1 Data

Abstract

Rice is considered the main food source for over 40% of the world population and plays a crucial role in countries’ food security, food management, and economic aspects. The value of SAR remote sensing in agricultural studies has its source of illumination and not limited to cloud cover. This makes it highly preferable over optical sensors in cloud-shrouded countries. The objective of the study is to assess the capability of Sentinel-1 data for determining paddy planting methods, identifying unhealthy paddy and an attempt made to differentiate rice varieties through correlation of in situ measurements and temporal variation of SAR backscattered signals. Six Sentinel-1 images are stacked to cover the entire paddy lifecycle. The correlated field data and plant backscatter showed that transplanted paddy has backscatter higher than broadcasted paddy. Two drops of paddy backscatter coefficient occurred, the first one, at the reproductive stage when paddy was attacked by bacteria and the second drop was at the ripening stage due to the attack of pests. The five rice varieties planted in Seberang Perak, Malaysia had the backscatter with insignificant differences that cannot confirm the Sentinel-1 capability to differentiate planted rice varieties. According to the obtained results, the time series of Sentinel-1 data has the capability for paddy rice growth monitoring.