中文摘要：

叶面积指数（LAI）是植物的重要结构变量,可以较好地反映植物冠层特征,构建该指标的有效反演模型是实现马尾松毛虫害空间监测与预警的必然要求。以在三明市、将乐县、沙县、南平市等4个县（市）测定的马尾松有效叶面积指数及同时段HJ-1 CCD多光谱遥感影像为基础数据,分别建立LAI与NDVI、TNDVI、DVI、RDVI、RVI、PVI、SAVI、MSAVI、MCARI等9个植被指数的一元线性、乘幂、指数、对数与二次曲线模型。结果表明：（1）上述9个植被指数均与马尾松有效叶面积指数显著相关;（2）R2、RMSE及预测精度等指标显示,以TNDVI与MSAVI作为自变量的乘幂、指数模型是马尾松毛虫主要寄主马尾松有效叶面积指数遥感反演的最佳模型,这与影像光谱信息量、指数对虫害状态下的林冠特征响应敏感有关。

英文摘要：

Leaf area index （LAI） is one of important structural variable of plant, and can effectively reflect plant canopy characteristics. To construct effective retrieving models for this indicator is inevitable requirement of spatial monitoring and early warning for Dendrolimus punctatus Walker damage. Taking the effective leaf area index of P. massoniana measured in 4 counties of Sanming, Jiangle, Shaxian and Nanping as well as same temporal H J-1 CCD multi-spectral images as the basic data, the linear, power, exponential, logarithmic and secondary curve models between LAI and 9 vegetation indices of NDVI、TNDVI、DVI、RDVI、RVI、PVI、SAVI、MSAVI、MCARI were constructed. The results show （1） there were significant correlations between these 9 vegetation indices and effective leaf area index of P. massoniana; （2） the statistical indicators of R^2, RMSE and estimation accuracy indicate that the power, exponential modes whose independent variables were TNDVI and MSAVI were the best effective leaf area index remote sensing retrieving models for the main host of Dendrolimus punctatus Walker-Pinus massoniana, this result was related to the spectral information amount of image and the sensitive response of the indices to canopy characteristics under pest status.