基于TerraSAR-X 和ALOS PALSAR 数据的森林蓄积量估测研究

Estimation of Forest Growing Stock Based on TerraSAR-X and ALOS PALSAR Data： a Case Study in Mengla County of Yunnan Province

摘要: 基于2014 年12 月3 日获取的X 波段TerraSAR-X 数据和2008 年10 月19 日获取的L 波段ALOS PALSAR 数据，引入树种类型为哑变量，采用逐步回归的方法，对云南省勐腊县森林蓄积量进行估测。结果表明，与X 波段TerraSAR-X 数据相比，基于L 波段ALOS PALSAR 数据建立的森林蓄积量模型具有更高的决定系数，R2 为0.843，模型估测精度为68.8%，均方根误差RMSE 为38.8 m3·hm-2，最终结果证明波长较长的L 波段ALOS PALSAR 数据对森林蓄积量具有更好的估测效果。

Abstract: X-band TerraSAR-X data of December 3rd of 2014 and L-band ALOS PALSAR data of October 19th of 2008 were applied for estimation of forest growing stock in Mengla county, Yunnan province by stepwise regression, taking tree species as dummy variable. Result showed that L-band ALOS PALSAR forest growing stock model had determination coefficient R2 of 0.843, accuracy of 68.8%, RMSE 38.8 m3/ha, better than X-band TerraSAR-X. It concluded that L-band ALOS PALSAR data had better estimation effect on forest growing stock.