GF-1/WFV在玉米叶面积指数估算中的应用研究

叶面积指数(leaf area index,LAI)是植被冠层重要的结构参数之一,与冠层生理过程密切相关,也是植被遥感领域关注的重要参数之一。本研究对已在轨运行7年的高分一号卫星WFV传感器的植被监测性能进行评测,以吉林省农安县典型玉米分布区作为研究区域,结合地面同步观测的叶面积指数和冠层光谱等实测数据,借助归一化植被指数(NDVI)、比植被指数(RVI)、大气阻抗植被指数(ARVI)、土壤调节植被指数(SAVI)、修改性土壤调节植被指数(MSAVI)这5种植被指数,对比分析地面实测光谱与GF-1/WFV光谱对玉米冠层叶面积指数的估算能力。通过决定系数(R²)、均方根误差(RMSE)、相对误差(RE)和预测残差(RPD)等参数筛选最优模型。研究结果显示,各种植被指数与LAI之间的相关性均表现为地面实测光谱高于GF-1/WFV星载光谱;对比不同植被指数与LAI的相关性发现,地面光谱和星上光谱构造的植被指数中,均表现为MSAVI与LAI的相关性最高;基于地面光谱和星上光谱的MSAVI构建的估算模型中,R²最高值所对应的函数类型不同,基于地面光谱的函数中,R²最高值对应的是指数模型,而基于GF-1/WFV星上光谱的函数中,二项式的R²最高。

Application research of GF-1/WFV data in estimation of maize leaf area index

The leaf area index (LAI), which plays a key role in response of the structural parameters of vegetation canopy, is closely related to the physiological process of the canopy. It is also one of the important parameters in the field of vegetation remote sensing. In the typical maize distribution area in Nong’an County of Changchun, combined with the measured data, the leaf area index and canopy spectrum were simultaneously observed on the ground. With the help of the normalized vegetation index (NDVI), ratio vegetation index (RVI), atmospheric regulation vegetation index (ARVI), soil adjustment vegetation index (SAVI), modified soil adjustment vegetation index (MSAVI).The ground measured spectrum data and GF-1/WFV spectrum data in estimating maize canopy leaf area index were compared and analyzed. After that, the optimal model was screened by parameters, such as coefficient of determination (R²), root-mean-square error (RMSE), relative error (RE) and residual prediction deviation (RPD). The results showed that the correlation between the different vegetation indexes and LAI of the ground-measured spectral data was higher than that of the GF-1/WFV spaceborne spectral data. In the comparison of the correlation between the different vegetation indexes and LAI, it was found that MSAVI had the highest correlation with LAI among the vegetation indexes of the ground spectrum and the satellite spectrum structure. In the estimation model based on ground spectral data and satellite spectral data, the function types corresponding to the highest R ² between MSAVI and LAI were different. For the measured spectrum, R ² was corresponded to the exponential function model, and for the satellite spectrum, it was corresponded to the quadratic function model.