Xu et al. (2026) A new adaptive approach for detecting crop sowing and harvesting dates from Sentinel-1 time series

Identification

• Journal: Remote Sensing of Environment
• Year: 2026
• Date: 2026-09-08
• Authors: Shuai Xu, Xiaolin Zhu
• DOI: 10.1016/j.rse.2026.115649

Research Groups

• Department of Land Surveying and Geospatial Science, The Hong Kong Polytechnic University
• Otto Poon Research Institute for Climate-Resilient Infrastructure, The Hong Kong Polytechnic University

Short Summary

This study proposes an adaptive Sentinel-1 Indicator Selection (ASIS) framework for detecting crop sowing and harvesting dates from Sentinel-1 time series. The framework integrates VH and CR signatures to reliably delineate crop growing seasons and detect CADs with high accuracy.

Objective

• Investigate the mechanisms influencing the performance of VH and CR indicators in cropping activity monitoring.
• Develop an adaptive SAR-based CADs detection framework capable of depicting field-scale CADs patterns that is robust to diverse crop types and agricultural management practices.

Study Configuration

• Spatial Scale: Field scale (3 km × 3 km buffer centered on PhenoCam camera coordinates)
• Temporal Scale: Daily temporal resolution, with annual Sentinel-1 VH and CR time series generated for each PhenoCam site

Methodology and Data

• Models used: Adaptive Sentinel-1 Indicator Selection (ASIS) framework, incorporating iterative time-series clustering and Radar Crop Growth Index (RCGI)
• Data sources: Sentinel-1 Ground Range Detected (GRD) products, PhenoCam Network data, ESA World Cover V100 product, Cropland Data Layer (CDL), European crop map, MYD10A1 Snow Cover Daily Global 500 m product

Main Results

• The ASIS framework achieved superior accuracy for sowing and harvesting dates estimation compared to using VH or CR alone.
• RCGI effectively integrated VH and CR signatures to reliably delineate crop growing seasons and detect CADs with high accuracy.

Contributions

• This study contributes an adaptive SAR-based CADs detection framework that is robust to diverse crop types and agricultural management practices.
• The proposed ASIS framework has the potential to improve crop modeling, field management, and yield estimation by providing accurate and timely information on CADs.

Funding

• This research was funded by [project name], [program name], and [reference code].

Citation

@article{Xu2026new,
  author = {Xu, Shuai and Zhu, Xiaolin},
  title = {A new adaptive approach for detecting crop sowing and harvesting dates from Sentinel-1 time series},
  journal = {Remote Sensing of Environment},
  year = {2026},
  doi = {10.1016/j.rse.2026.115649},
  url = {https://doi.org/10.1016/j.rse.2026.115649}
}