Co-Registration of Two DEMs: Impacts on Forest Height Estimation From SRTM and NED at Mountainous Areas
作者:Ni, WJ (Ni, Wenjian)[ 1 ] ; Sun, GQ (Sun, Guoqing)[ 2 ] ; Zhang, ZY (Zhang, Zhiyu)[ 1 ] ; Guo, ZF (Guo, Zhifeng)[ 1 ] ; He, YT (He, Yating)[ 3 ]
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
卷: 11  期: 1  页: 273-277
DOI: 10.1109/LGRS.2013.2255580
出版年: JAN 2014

摘要
The digital elevation model from the Shuttle Radar Topography Mission (SRTM) and the National Elevation Dataset (NED) have been used to estimate the forest canopy height. Most of such studies have been conducted over flat areas; the method performance has not been carefully examined over mountainous areas. This study, which is conducted over two mountainous test sites located in California and New Hampshire, reveals that the co-registration of these two digital elevation models (DEMs) is crucial to ensuring the quality of the results. The image co-registration method used in interferometric SAR processing is adapted to the co-registration of two DEMs. The forest canopy height from the Laser Vegetation Imaging Sensor (LVIS) is used as the reference data. The results showed that the misregistration between SRTM and NED was very obvious at both test sites. After the co-registration, the R-2 of the correlation between the height of the C-band scattering phase center derived from SRTM minus NED and the forest canopy height derived from LVIS data was improved from 0.19 to 0.51, and RMSE was reduced from 16.4 m to 6.8 m for slope up to 55 degrees at the California test site, while the R-2 was improved from 0.39 to 0.57 and RMSE was reduced from 5.4 m to 3.6 m for slopes up to 45 degrees at the New Hampshire test site. The influences of data resolution and terrain slopes were also investigated. The results showed that reducing the data resolution by spatial averaging could not reduce the influence of DEM misregistration.

通讯作者地址: Ni, WJ (通讯作者)
Jointly Sponsored Inst Remote Sensing Applicat Ch, State Key Lab Remote Sensing Sci, Beijing 100101, Peoples R China.
地址:
[ 1 ] Jointly Sponsored Inst Remote Sensing Applicat Ch, State Key Lab Remote Sensing Sci, Beijing 100101, Peoples R China
[ 2 ] Univ Maryland, College Pk, MD 20742 USA
[ 3 ] Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Beijing 100101, Peoples R China