以福建中部的三明市、沙县和将乐县以及福建西部的长汀县为研究区,利用郁闭度(CD)与HJ-1 多光谱数据(B₁ 、B₂ 、B₃ 和B₄ )并结合实地踏查数据,对马尾松(Pinus massoniana Lamb.)林叶面积指数(LAI)进行估算。结果显示:利用多光谱波段及归一化差值植被指数(NDVI),得到三明市、沙县和将乐县马尾松林郁闭度估算模型为CD=-0. 008B₁ +0. 031B₂ +0. 016B₃ -0. 037B₄ +0. 012NDVI-0. 815,平均检验精度为87. 76%;长汀县马尾松林的郁闭度估算模型为CD =-0. 024B₁ +0. 130B₃ -0. 083B₄ +0. 026NDVI-2. 355,平均检验精度为86. 78%。采用线性、对数、二次、幂和指数模型构建了2 个研究区马尾松林的LAI-CD 模型,经检验分析,均以指数模型综合预测效果最佳;综合分析后获得研究区马尾松林LAI-CD 模型LAI =0. 193e^{3. 259CD} ,检验精度达0. 898。研究结果说明:利用国产HJ-1多光谱数据可较精确地估算马尾松林郁闭度,所构建的LAI-CD 数学模型预测效果较好,具有较强的适用性。

Taking Sanming City, Shaxian County and Jiangle County in Central Fujian Province and Changting County in West Fujian Province as research areas, leaf area index (LAI) of Pinus massoniana Lamb. forest was estimated by using of canopy density (CD), HJ-1 multi-spectral data (B1, B2, B3 and B4) and combining with investigation data in field. The results show that by using multi-spectral bands and normalized difference vegetation index ( NDVI ), the estimated model of canopy density of P .massoniana forest in Sanming City, Shaxian County and Jiangle County is obtained, that is CD=-0. 008B₁ +0. 031B₂ +0. 016B₃ -0. 037B₄ +0. 012NDVI-0. 815, with the average estimating accuracy of87. 76%; and the model of P. massoniana forest in Changting County is CD =-0. 024B₁ +0. 130B₃ -0. 083B₄ +0. 026NDVI-2. 355, with the average estimating accuracy of 86. 78%. The models of LAI-CDof P. massoniana forest in two research areas are constructed by using of linear, logarithmic, quadratic, power and exponential models, and the verifying analysis indicates that the comprehensive estimatingeffect of exponential model is the best. According to comprehensive analysis result, the LAI-CD model LAI = 0. 193e^{3. 259CD} of P. massoniana forest in two research areas was obtained with the estimating accuracy of 0. 898. It is suggested that canopy density of P. massoniana forest can be estimated accurately by means of China’s domestic HJ-1 multi-spectral data, and the LAI-CD model established possesses better prediction effect and stronger applicability.