基于NIR-Red光谱特征空间的作物水分指数

水分含量是表征作物水分胁迫生理状况的重要指标，及时有效地监测作物水分含量对于评估作物水分亏缺平衡，指导农业生产灌溉具有重要意义。基于NIR-Red二维光谱特征空间，尝试构建一种新的作物水分监测指数PWI来估算作物水分含量。以冬小麦作物植被水分含量估算为尝试对象。首先，利用地面实测小麦冠层高光谱数据，结合对应卫星光谱响应函数，模拟当前常用卫星HJ-CCD和ZY-3多光谱数据;然后，对基于NIR-Red二维光谱特征空间的现有植被指数PDI(垂直干旱指数)和PVI(垂直植被指数)进行改进，通过比值变换的方法构建新的指数PWI来估算冬小麦植株含水量(VWC)。结果显示：基于模拟的HJ-CCD和ZY-3卫星宽波段多光谱数据生成的PWI估算小麦VWC具有良好的效果，R2分别达到0.684和0.683, 均达到了极显著水平。利用检验样本得到冬小麦VWC估算的R2和RMSE分别为0.764和0.764，3.837%和3.840%，这表明应用提出的新指数PWI估测作物含水量具有一定可行性。同时，也为当前利用主要国产卫星遥感数据HJ-CCD和ZY-3开展作物水分遥感监测应用提供了一种新方法。

The New Method Monitoring Crop Water Content Based on NIR-Red Spectrum Feature Space

Moisture content is an important index of crop water stress condition, timely and effective monitoring of crop water content is of great significance for evaluating crop water deficit balance and guiding agriculture irrigation. The present paper was trying to build a new crop water index for winter wheat vegetation water content based on NIR-Red spectral space. Firstly, canopy spectrums of winter wheat with narrow-band were resampled according to relative spectral response function of HJ-CCD and ZY-3. Then, a new index (PWI) was set up to estimate vegetation water content of winter wheat by improveing PDI (perpendicular drought index) and PVI (perpendicular vegetation index) based on NIR-Red spectral feature space. The results showed that the relationship between PWI and VWC(vegetation water content) was stable based on simulation of wide-band multispectral data HJ-CCD and ZY-3 with R2 being 0.684 and 0.683, respectively. And then VWC was estimated by using PWI with the R2 and RMSE being 0.764 and 0.764, 3.837% and 3.840%, respectively. The results indicated that PWI has certain feasibility to estimate crop water content. At the same time, it provides a new method for monitoring crop water content using remote sensing data HJ-CCD and ZY-3.