Tan Pengyuan, Zhu Jianjun, Fu Haiqiang, Wang Changcheng, Liu Zhiwei, Zhang Chen
School of Geosciences and Info-Physics, Central South University, Changsha 410083, China.
The Second Monitoring and Application Center, China Earthquake Administration, Xi'an 710054, China.
Sensors (Basel). 2020 Dec 19;20(24):7304. doi: 10.3390/s20247304.
This paper develops a framework for extracting sub-canopy topography from the TanDEM-X digital elevation model (DEM) by fusing ALOS-2 PARSAR-2 interferometric synthetic aperture radar (InSAR) coherence and Global Ecosystem Dynamics Investigation (GEDI) data. The main idea of this method is to estimate the forest height signals caused by the limited penetration of the X-band into the canopy from the TanDEM-X DEM. To achieve this goal, a spaceborne repeat-pass InSAR coherent scattering model is first used to estimate the forest height by the ALOS-2 PARSAR-2 InSAR coherence (APIC), taking the GEDI canopy height as the reference. Then, a linear regression model of the TanDEM-X DEM Vegetation Bias (TDVB) depending on the forest height and the fraction of vegetation cover (FVC) is established and used to estimate the sub-canopy topography. The proposed method was validated by the data of the Amazon rainforest and a boreal forest in Canada. The results showed that the proposed method extracted the sub-canopy topography at the study sites in the tropical forest and boreal forest with the root mean square error of 4.0 m and 6.33 m, respectively, and improved the TanDEM-X DEM accuracy by 75.7% and 39.7%, respectively.
本文通过融合ALOS-2 PARSAR-2干涉合成孔径雷达（InSAR）相干性和全球生态系统动力学调查（GEDI）数据，开发了一种从TanDEM-X数字高程模型（DEM）中提取林下地形的框架。该方法的主要思想是从TanDEM-X DEM中估计由X波段对树冠穿透有限所引起的森林高度信号。为实现这一目标，首先利用星载重复轨道InSAR相干散射模型，以GEDI树冠高度为参考，通过ALOS-2 PARSAR-2 InSAR相干性（APIC）来估计森林高度。然后，建立了一个依赖于森林高度和植被覆盖度（FVC）的TanDEM-X DEM植被偏差（TDVB）线性回归模型，并用于估计林下地形。所提方法通过亚马逊雨林和加拿大北方森林的数据进行了验证。结果表明，所提方法在热带森林和北方森林的研究地点分别以4.0米和6.33米的均方根误差提取了林下地形，并且分别将TanDEM-X DEM的精度提高了75.7%和39.7%。
