2014-2023年黄淮海平原冬小麦单产数据集
收藏资源简介:
基于现有冬小麦单产数据集空间分辨率较低问题,整合作物生长模型和深度学习模型,生产了2014—2023年冬小麦20m/30m单产数据集(HHHWheatYield20m/30m)。该数据在2014—2016年使用Landsat-8数据,分辨率为30m,2017—2023年使用Sentinel-2数据,分辨率为20m。该数据与县级统计数据相比MRE为6.63%,与实测数据相比,MRE为11.83%。该数据可为评价黄淮海平原冬小麦单产提供数据支撑。数据格式为GeoTIFF,单位为千克/公顷(kg/ha)。
Aiming at the low spatial resolution issue of existing winter wheat yield datasets, this study integrates crop growth models and deep learning models to develop the 2014–2023 winter wheat yield dataset (HHHWheatYield20m/30m). Specifically, Landsat-8 data with a 30 m spatial resolution was used for the 2014–2016 period, while Sentinel-2 data with a 20 m spatial resolution was adopted for 2017–2023. Compared with county-level statistical yield data, the Mean Relative Error (MRE) of this dataset is 6.63%, and compared with in-situ measured yield data, the MRE reaches 11.83%. This dataset can provide robust data support for evaluating winter wheat yield in the Huang-Huai-Hai Plain. The dataset is stored in GeoTIFF format, with the unit of kilograms per hectare (kg/ha).




