华北农业干旱监测与冬小麦估产研究

评估了基于植被光学厚度(VOD)和日光诱导叶绿素荧光(SIF)等植被广义光学特性构建的标准化植被指数(Z_VI)监测农业干旱的适用性；并采用VOD、SIF两种指数和土壤水分等环境变量的不同组合建立冬小麦估产的岭回归模型，以探究其对冬小麦产量的预报能力。结果表明：相较于Z_VOD,旬尺度Z_SIF对华北地区的农业干旱具有更好的监测能力，对重旱的正确检测率达到77%。Z_SIF能够有效反映干旱发生、发展直至减轻的演变过程，其低值区与站点记录的干旱空间分布相吻合。在华北地区南部，生长季C波段和Ku波段的VOD对冬小麦单产的预报能力优于SIF;利用VOD、SIF两种指数和环境变量的全变量模型取得了最好的估产精度，影响冬小麦产量估算精度的关键预测变量为生长高峰期的SIF。研究可为大范围农业干旱监测和粮食安全提供技术支持。

Agricultural drought monitoring and winter wheat yield estimation in North China

This study evaluates the applicability of standardized vegetation index (Z_VI) constructed by generalized optical properties of vegetation including Vegetation Optical Depth (VOD) and Sun/Solar-induced Chlorophyll Fluorescence (SIF) to monitor agricultural drought, and further explores the prediction ability of VOD, SIF and the environmental variables including soil moisture on winter wheat yield based on ridge regression model. The results show that the ten-day scale Z_SIF outperforms Z_VOD in the agricultural drought monitoring ability in North China, with the probability of detection (POD) for severe drought reached 77%. Z_SIF can effectively reflect the evolution process of drought occurrence, development and mitigation, and its low-value area is consistent with the spatial distribution of stations recording drought occurrences. In the south of North China, the ability of VOD in C-band and Ku-band during growing season is higher than SIF for winter wheat yield estimation. The highest estimation accuracy was obtained by using the full-feature model including VOD, SIF and the environmental variables. The crucial predictor of winter wheat yield is SIF at the growth peak. This study can provide technical support for large-scale agricultural drought monitoring and food security.