利用Beer-Lambert消光定律遥感反演玉米叶面积指数

Retrieval of Corn Leaf Area Index Using Beer-Lambert Extinction Law

利用遥感技术反演大范围玉米叶面积指数,对于田间肥水管理、长势监测乃至产量预测具有重要意义。在野外实测样本的支持下,获取玉米抽雄期的Landsat-8多光谱影像,引进Beer-Lambert定律,利用最小二乘法分析玉米冠层结构的消光系数,构建玉米叶面积指数遥感反演模型,最后采用交叉验证法评价模型精度。结果表明:玉米抽雄期NDVI、LAI呈较明显的正相关关系;基于Beer-Lambert定律的玉米叶面积指数遥感反演模型决定系数可达0.97,LAI空间分布状况与当地农业技术推广部门掌握的玉米实际生长状况基本一致,说明利用Beer-Lambert消光定律方法可以有效地反映玉米群体结构对光照的影响,据此开展玉米叶面积指数遥感反演具有较高的可行性。

Using remote sensing technology to retrieve a large area of maize leaf area index is of important significance for field fertilizer and water management,growth monitoring and even yield forecasting.In the support of field survey samples,this paper obtained the Landsat-8multispectral images of maize tasseled during the tasseling stage, introduced Beer-Lambert law,analyzed the extinction coefficient of corn canopy structure by least square method,and constructed a remote sensing inversion model of maize leaf area index.Finally,the cross validation method was used to evaluate the accuracy of the model.The results showed that there was a significant positive correlation between NDVI and LAI in maize tasseling stage.Based on the Beer-Lambert law,the determination coefficient of maize leaf area index was 0.97.The spatial distribution of LAI was basically consistent with the actual maize growth situation mastered by local agricultural extension department,indicating that using the Beer-Lambert extinction law could effectively reflect the effect of population structure of maize on illumination.Hereby,it was highly feasible to carry out remote sensing retrieval of maize leaf area index.