不同生育期冬小麦叶面积指数高光谱遥感估算模型

发布时间：2018-12-14 08:39

【摘要】：基于地面实测的冬小麦的生理生态参数数据和冠层光谱数据,分析返青期、拔节期、抽穗期、开花期冬小麦叶面积指数与原始光谱及其一阶微分的相关性,并构建基于等效TM数据的植被指数,建立不同生育时期的冬小麦叶面积指数(LAI)的高光谱遥感估算模型。结果表明:(1)返青期、拔节期、抽穗期的冬小麦LAI与原始光谱相关性较好,在400~720 nm波长范围内呈负相关,在720~900 nm之间呈正相关,开花期的冬小麦LAI与冠层光谱相关性较差;(2)返青期、拔节期冬小麦LAI与光谱一阶微分显著相关,分别在480~540 nm、550~580 nm形成波峰、波谷,在670~760 nm范围内形成"平台",相关系数达到0.8以上,但抽穗期、开花期LAI与光谱一阶微分的相关性较差;(3)在等效植被指数与返青期、拔节期和抽穗期LAI建立的回归模型中,分别使用m SRI、RVI与MSAVI2建立的幂函数模型或指数模型最佳,最优模型分别为y=0.053e4.962x,y=0.409x0.828,y=18.687x3.061,对应的r2分别为0.589、0.648、0.694,开花期不适宜使用等效植被指数建立遥感监测模型。
[Abstract]:Based on the data of physiological and ecological parameters and canopy spectrum of winter wheat measured on the ground, the correlation between the leaf area index of winter wheat and its original spectrum and its first order differential was analyzed at the stage of turning green, jointing, heading and flowering. Based on the equivalent TM data, a hyperspectral remote sensing estimation model of winter wheat leaf area index (LAI) was established. The results showed that: (1) the LAI of winter wheat at the stage of turning green, jointing and heading had a good correlation with the original spectrum, with a negative correlation in the wavelength range of 400-720 nm, a positive correlation between 720 nm and 900 nm, and a poor correlation between the LAI and the canopy spectrum at flowering stage; (2) there was a significant correlation between LAI and the first order differential spectrum of winter wheat at green and jointing stage. The peaks and troughs were formed at 480 ~ 540 nm,550~580 nm, and the "platform" was formed in the range of 670 ~ 760 nm. The correlation coefficient was more than 0. 8, but at heading stage. The correlation between LAI and the first order differential spectrum was poor at anthesis. (3) among the regression models established by LAI in equivalent vegetation index and green stage, jointing stage and heading stage, the power function model or exponential model established by m SRI,RVI and MSAVI2 respectively is the best. The corresponding R2 is 0.589U 0.6480.694.It is not suitable to use the equivalent vegetation index to establish the remote sensing monitoring model in flowering period.
【作者单位】：山东省农业可持续发展研究所;中国矿业大学资源与地球科学学院;
【基金】：国家自然科学基金(编号:41401407) 山东省自然科学基金(编号:ZR2014YL016)
【分类号】：S127;S512.11

【相似文献】