估测可燃物载量的遥感及地形因子选择探究

摘要

In order to improve the estimation accuracy of forest fuel load , we should select the combination of remote sensing factors and Topographic factors that has smaller correlation. In this paper with the residual sum of squares method, we set up in the Tahe County, Heilongjiang Province, 96.5 km northwest of Pangu forest plots, surveying the sample measured factors of forest fuel load. Combined with the measured value ,we search the correlation com- bination of remote sensing factors and topographic factors on the impact of the fuel load. The results show that de- spite the remote sensing factors RVI, DVI, SARVI, TSAVI and Topographic factors ASPECT, DEM, SLOPE, the remaining factors have little effect in the estimation of forest fuel loads: Removing SARVI and ASPECT, the multiple correlation between each factor significantly reduce the variance to expand the analysis of the factors of each factor.%为提高森林可燃物载量的估测精度,应选择相关性小的遥感因子及地形因子组合。本文运用残差平方和法,在黑龙江省塔河县西北部96.5km处的盘古林场设立样地,调查样地内可燃物载量实测因子,结合实测值对影响可燃物载量的遥感及地形因子的组合进行了相关性研究。结果表明除遥感因子RVI、DVI、SARVI、TSAVI及地形因子ASPECT、DEM、SLOPE外,其余因子对森林可燃物载量的估测影响不大。通过方差扩大因子对各因子的相关性分析表明：剔除SARVI及ASPECT后,各因子间的多重相关性大幅减小。