Abstract
Crop rotations, the farming practice of growing crops in sequential seasons, occupy a core position in agriculture management, showing a key influence on food security and agro-ecosystem sustainability. Despite the improvement in accuracy of identifying mono-agricultural crop distribution, crop rotation patterns remain poorly mapped. In this study, a hybrid convolutional neural network (CNN) and long short-term memory (LSTM) architecture, namely crop rotation mapping (CRM), were proposed to synergize the synthetic aperture radar (SAR) and optical time series in a rotational mapping task. The proposed end-to-end architecture had reasonable accuracies (i.e., accuracy > 0.85) in mapping crop rotation, which outperformed other state-of-the-art non-deep or deep-learning solutions. For some confusing rotation types, such as fallow-single rice and crayfish-single rice, CRM showed substantial improvements from traditional methods. Furthermore, the deeply synergistic SAR-optical, time-series data, with a corresponding attention mechanism, were effective in extracting crop rotation features, with an overall gain of accuracy of four points compared with ablation models. Therefore, our proposed method added wisdom to dynamic crop rotation mapping and yields important information for the agro-ecosystem management of the study area.
Highlights
Crop rotation is the sequential growth of crops in a year [1] and has been practiced for thousands of years, serving as one of the most effective ways to increase grain yields [2,3].a crop rotation system is believed to have profound advantages over a monoculture cropping system for improving cropland fertility and mitigating weeds and pest insects [4]
Showed a slight advantage over NDVI and enhanced vegetation index (EVI) only when used with crop rotation mapping (CRM), and achieved the highest accuracy of 0.876 among all models
NDVI achieved an overall stable and preferable performance in the rotation mapping task and provided the best results in the two basic competing models random forest (RF)(S2) and long short-term memory (LSTM)(S2), where there was a large gap between EVI and two kinds of NDVI
Summary
Crop rotation is the sequential growth of crops in a year [1] and has been practiced for thousands of years, serving as one of the most effective ways to increase grain yields [2,3].a crop rotation system is believed to have profound advantages over a monoculture cropping system for improving cropland fertility and mitigating weeds and pest insects [4]. Temporal data on crop types and crop rotations at the field level for regional scales are rarely available, preventing further spatial explicit assessment and analysis [6,7]. Most regional or global studies have to rely on the assumption of a few simplified prototype crop rotations, bringing large uncertainty to evaluation results [4]. There have been a rather limited number of studies that attempted to derive crop rotation maps compared to the conventional single-season crop mapping. Most studies have generated rotation maps based on spatial overlay analysis of singleseason crop mapping results [8,9], i.e., to integrate rotation types by overlaying separate
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
