Detection of Spatio-Temporal Rice Cropping Systems in the Mekong Delta Using LongTerm MODIS Time-Series Data

Abstract

In the Vietnamese Mekong Delta (VMD), where frequent flooding, salinity and drought affect rice productivity, case studies of comprehensive spatio-temporal detection systems for long-term geospatial research are lacking. Thus, we developed a rule-based high-accuracy algorithm, termed Rice Phenology Detection System (RicePDS), for estimating spatio-temporal information on cropping intensity, sowing dates, and flowering dates of paddy fields using Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data (2002–2020). We based these estimations on the vegetation, flood, surface temperature, and cloud signals of rice cropping systems in the VMD. RicePDS could be useful for deriving the spatio-temporal distribution of cropping intensity, rice area, and durations of sowing and flowering periods for analysis. The estimated paddy rice area and multi-year agricultural statistical trends in rice cultivation were highly accurate.