文献类型：期刊文献

英文题名：High-resolution subcanopy topography inversion for boreal forest areas based on an IWCM-aided multi-baseline InSAR method

作者：Wang, Huiqiang[1,2] Feng, Zhimin[1,2] Fu, Haiqiang[3] Zhu, Jianjun[3] Li, Ruiping[1,2] Liu, Zhiwei[4] Wang, Lei[5] Hao, Shuai[6,7] Zhang, Qiuliang[6,7] Zhao, Lei[8] Zhao, Zheng[9]

第一作者：Wang, Huiqiang

通信作者：Fu, HQ[1]

机构：[1]Inner Mongolia Agr Univ, Coll Water Conservancy & Civil Engn, Hohhot 010018, Peoples R China;[2]Inner Mongolia Agr Univ, State Key Lab Water Engn Ecol & Environm Arid Area, Hohhot 010018, Peoples R China;[3]Cent South Univ, Sch Geosci & Info Phys, Changsha 410083, Peoples R China;[4]Cent South Univ Forestry & Technol, Coll Adv Interdisciplinary Studies, Changsha 410004, Peoples R China;[5]Southwest Petr Univ, Sch Civil Engn & Geomatics, Chengdu 640500, Peoples R China;[6]Inner Mongolia Agr Univ, Forestry Coll, Hohhot 010019, Peoples R China;[7]Forest Ecosyst Natl Observat & Res Stn Greater Khi, Genhe 022350, Peoples R China;[8]Chinese Acad Forestry, Inst Forest Resource Informat Tech, 1 Dongxiaofu, Beijing 100091, Peoples R China;[9]Chinese Acad Surveying & Mapping, Beijing 100036, Peoples R China

年份：2025

卷号：230

起止页码：547-562

外文期刊名：ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING

收录：;EI(收录号：20254119283385);WOS:【SCI-EXPANDED(收录号:WOS:001592856700001)】；

基金：This work is supported by the National Key Research and Develop-ment Program of China (No. 2022YFB3902605) , National Natural Sci-ence Foundation of China (42204024) , High-level/excellent doctoral talents introduction research project in Inner Mongolia Agricultural University (NDYB2022-27) , and Natural Science Foundation project of Inner Mongolia (2024MS04006) . The authors would also like to acknowledge the German Aerospace Centre (DLR) for providing the TanDEM-X data over study areas (ID. NTI_BIST7299) , and the National Snow and Ice Data Center for providing the ICESAT-2 data, and the European Space Agency (ESA) for providing the Sentinel-2 data, and the Swedish Defense Research Agency (FOI) for providing the LiDAR data.

语种：英文

外文关键词：Subcanopy topography; Interferometric synthetic aperture radar (InSAR); Interferometric water cloud model (IWCM); Scattering phase center (SPC) height; Boreal forest areas; Ground-to-volume amplitude ratio

摘要：The extraction of the large-scale, high-resolution (better than 10 m) subcanopy topography in forested areas has been a significant challenge. At present, bistatic and formation-flying interferometric synthetic aperture radar (InSAR) techniques have been applied to forest canopy height inversion and topography mapping using empirical models (e.g., SINC model, penetration depth model). These approaches only account for the volume scattering process and are inadequate for discontinuous boreal forests where the X-band SAR signal can penetrate canopy gaps to reach the ground surface. To address these limitations, this study introduces an interferometric water cloud model (IWCM) that simultaneously models both volume scattering and surface scattering without requiring external forest height data. However, the model brings new problems: six unknown parameters to be estimated, high nonlinearity, and difficult convergence in parameter estimation. To solve these problems, the normalized difference vegetation index (NDVI) is used to characterize the forest density in the horizontal direction, and the SAR amplitude information is applied to further quantify the ground-to-volume amplitude ratio. Further, a nonlinear least squares (non-LS) iterative algorithm is utilized during parameter estimation. Based on the above, this study proposes a framework of IWCM-aided multi-baseline InSAR inversion to estimate the scattering phase center (SPC) height and the subcanopy topography. We validate the method using multi-baseline TanDEM-X data from the German Aerospace Center (DLR) over two boreal forest sites, the Krycklan test site in northern Sweden and the Genhe test site in northeast China, with ICESAT-2 and airborne LiDAR data as reference. The airborne LiDAR assessment demonstrates the extracted subcanopy topography with the ME of 1.44 m (Krycklan) and 1.01 m (Genhe), RMSEs of 3.72 m and 4.0 m, respectively. It seems that a systematic bias is present in subcanopy topography inversion. But in fact, the systematic bias can be linked to the influences of critical factors, including NDVI, the ratio (m(0)) of the backscattering amplitude from the ground surface (sigma(gr)) to that from pure forest layer (sigma(veg)), the multi-baseline number and geometric configuration. After the investigations, it is confirmed that the bias can be eliminated or significantly reduced by some reasonable strategies, suggesting the improvement of inversion accuracy. With the advent of high-resolution formation-flying SAR constellations like China's Hongtu-1, this framework presents a promising solution for high-precision subcanopy topographic mapping in boreal forests.