冬小麦病害与产量损失的多时相遥感监测

Monitoring and evaluation of the diseases of and yield winter wheat from multi-temporal remotely-sensed data

为了开展农作物病害遥感监测与产量损失评估,该文以北京郊区大田生产条件下的冬小麦条锈病、白粉病的为研究对象,获取了2007年4月10日、4月26日、5月12日、5月28日共四期Landsat TM卫星影像,准同步地测量了试验地块的冠层光谱数据及配套农学数据。利用该4个时相的遥感数据,分析了试验区的冬小麦条锈病、白粉病在主要生育期的光谱特征及其变化,与对照地块相比,病害小麦在可见光和短波红外波段的光谱反射率降增大,近红外波段反射率减小,红边则会向短波方向移动,红边振幅减小,NDVI值减小。并利用冬小麦病害发生前期(4月10日,4月26日)的卫星遥感数据建立了作物产量的早期预测模型,结合实测的产量数据,定量计算了条锈病和白粉病的产量损失,结果表明两个白粉病和条锈病小麦地块的减产幅度超过了30%。

Remote sensing techniques can provide crop growth information economically, rapidly, and objectively on a large scale., It has been widely used to monitor crop diseases. In this study, four Landsat TM images were acquired in the Beijing areas in winter wheat at four stages of growth---erecting stage （Apr. 10, 2007）, booting stage （Apr. 26, 2007）, anthesis stage （May 12, 2007） and grain filling stage （May 28, 2007） in Beijing. At the first three stages, the relative canopy spectra were also measured. All these multi-temporal remotely sensed data were analyzed to detect and evaluate winter wheat rust stripe and powdery mildew diseases. If diseases occurred, the spectral reflectances of diseased crops at visible and short infrared regions decreased, and the spectral reflectances at near-infrared region increased due to the decrease of plant chlorophyll concentration, water content and leaf area index （LAI）; In addition, the red edge position decreased, namely, ＂blue shift＂ and NDVI values also decreased. Moreover, an early yield prediction model was developed using the Landsat TM images acquired at the erecting and booting stages, and the yield loss due to diseases was evaluated by using the measured yield and the predicted yield.