Recently, the Innovation Team of Smart Agriculture of the Institute of Agricultural Resources and Regional Planning (IARRP) of the Chinese Academy of Agricultural Sciences (CAAS), in collaboration with leading domestic institutions, have introduced a new method for generating samples and mapping single and double cropping rice by synergizing optical and microwave data. The related findings have been published in the flagship journal of the remote sensing field, "Remote Sensing of Environment" (IF=11.1).

Accurately capturing the spatiotemporal distribution information of rice planting patterns is of significant importance for yield estimation, water resource management, and greenhouse effect assessment. However, existing high-precision mapping methods often rely on abundant and reliable training samples and high-quality temporal remote sensing data unaffected by cloud cover or rainfall.

The research team has developed a method called POSTAR (Phenology domain Optical-SAR feature inTegration method to Automatically generate single and double cropping Rice sample), based on the synergistic optical-microwave features in the phe nology domain, which has successfully generated high-quality samples of single and double cropping rice. Compared to field samples, the spectral correlation similarity is high (>0.85), and the dynamic time warping distance is small (<8.5). Applying POSTAR-generated samples and a random forest classifier, the overall accuracy of spatial distribution of single and double cropping rice in the Dongting Lake Plain reached 89.6%, with F1 scores of 0.899 and 0.938 for single cropping rice and double cropping rice, respectively. In the Poyang Lake Plain and Sanjiang Plain, the overall classification accuracies were 96.0% and 97.8%, respectively. This study innovatively combines optical and SAR data, providing a valuable perspective for large-scale crop monitoring in complex agricultural conditions with frequent cloud cover and rainfall.

Dr. Yang Jingya from IARRP is the first author of the paper, with Researcher Wu Wenbin and Associate Professor Hu Qiong from Central China Normal University serving as the corresponding authors. The research received funding support from the National Key Laboratory for Efficient Utilization of Dryland Cropland in Northern China, the National Key Research and Development Program (2019YFA0607400), National Natural Science Foundation of China projects (42271399 and 41921001), Central University Basic Research Business Special Fund (CCNU24JCPT012, CCNU24JCPT031), and Xinjiang Uygur Autonomous Region Science and Technology Project (2022LQ02004).

Phenological Characteristics Curve of Single and Double Cropping Rice in the Dongting Lake Plain

Original Article Link: https://doi.org/10.1016/j.rse.2024.114387