论文部分内容阅读
由于测定树木胸径、树高等参数较为困难,使得地上生物量的时空估计成为一项很困难的任务。本研究在印度兰契市的Birla技术学院进行,该校区长满娑罗双树。野外测定的地上生物量分别与遥感数据的个体条带、条带主成分、植被指数、植被指数主成分进行线性回归分析,以及地上生物量分别与这些参数进行多元线性回归分析,决定地上生物量与遥感参数之间关联性。线性回归分析表明,只有NDVI回归系数值在0.8以上,其他参数的回归系数值均较低。另外,多元线性回归方程计算得到的地上生物量与野外测定的值的相关系数在0.9以上,说明用多元线性回归法估计地上生物量具有更好的可靠性。多元回归分析中植被指数主成分与地上生物量之间的相关性系数是0.99。
It is difficult to estimate the spatio-temporal biomass of above-ground biomass because of the determination of DBH and tree height parameters. The study was conducted at Birla Institute of Technology in Ranchi, India, which is full of Milo trees. The aboveground biomass was determined by linear regression analysis with the individual bands, the main components of the bands, the vegetation index and the vegetation index of the remote sensing data, and the multiple linear regression analysis of the aboveground biomass with these parameters respectively to determine the aboveground biomass And remote sensing parameters. Linear regression analysis showed that only the NDVI regression coefficients were above 0.8, and the regression coefficients of other parameters were lower. In addition, the correlation coefficient between above-ground biomass calculated by multivariate linear regression equation and field measurements was above 0.9, indicating that multivariate linear regression method has better reliability in estimating aboveground biomass. In the multiple regression analysis, the correlation coefficient between vegetation index and aboveground biomass was 0.99.