Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Water Management and Systems Research » Research » Publications at this Location » Publication #369906
Research Project: Improving the Sustainability of Irrigated Farming Systems in Semi-Arid Regions

Location: Water Management and Systems Research

Title: Winter wheat mapping based on Sentinel-2 data in heterogeneous planting conditions

Author
item ZHANG, DONGYAN - Anhui Agricultural University
item FANG, SHENMEI - Anhui Agricultural University
item SHE, BAO - Anhui Agricultural University
item Zhang, Huihui
item JIN NING - Anhui Agricultural University
item XIA, HAOMING - Henan University
item YANG, YUYING - Anhui Agricultural University
item DING, YANG - Anhui Agricultural University

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/11/2019
Publication Date: 11/13/2019
Citation: Zhang, D., Fang, S., She, B., Zhang, H., Jin Ning, Xia, H., Yang, Y., Ding, Y. 2019. Winter wheat mapping based on Sentinel-2 data in heterogeneous planting conditions. Remote Sensing. 11(22):2647. https://doi.org/10.3390/rs11222647.
DOI: https://doi.org/10.3390/rs11222647

Interpretive Summary: Monitoring and mapping spatial distribution of winter wheat (Triticum Aestivum L.) plantation accurately became important for crop management, damage assessment and yield prediction in Anhui province, China. In this study northern Anhui counties and central Anhui counties were selected as study areas, and Sentinel-2 imagery was used to map winter wheat distribution in the 2017-2018 growing season. The Sentinel-2 imagery at the heading stage was identified as the optimum period for winter wheat area extraction. Spectral wavebands and vegetation indices (VI) generated from the images at the heading stage were used to map winter wheat areas. The result showed better mapping accuracy for northern Anhui than central Anhui due to the difference in field topography and land surface.

Technical Abstract: Monitoring and mapping spatial distribution of winter wheat (Triticum Aestivum L.) accurately became important for crop management, damage assessment and yield prediction in Anhui province, China. In this study northern Anhui counties and central Anhui counties were selected as study areas, and Sentinel-2 imagery was employed to map winter wheat distribution and the results were verified with Planet imagery in the 2017-2018 growing season. The Sentinel-2 imagery at the heading stage was identified as the optimum period for winter wheat area extraction. Therefore, ten spectral features and nine vegetation indices (VI) generated from the images at the heading stage were used to classify winter wheat areas by Random Forest algorithm. The result showed the accuracy was between 93% and 97%, with Kappa above 0.82 and percentage error (PE) lower than 5% in northern Anhui, and the accuracy about 80% with Kappa ranging from 0.70 - 0.78 and PE about 20% in central Anhui. Northern Anhui had large planting scale of winter wheat and flat terrain while central Anhui had relatively small winter wheat fields and high degree of surface fragmentation, which made the extraction effect in central Anhui inferior to that in northern Anhui area. Further, the results obtained by using the optimum subset data from VIs and spectral bands were better than other datasets with the great advantage in data volume and processing time. This study provides a perspective for winter wheat mapping under various climatic and complicated land surface conditions and is of great significance for crop monitoring and agricultural decision-making.