Reduction of errors in SRTM DEM using ICESat lidar altimetry

DOI

Bibliographic Information

Other Title
  • ICESatレーザー高度計を用いたSRTM DEMの高精度化

Description

In this paper, we introduce a method for removing height errors from the SRTM DEM using ICESat lidar altimetry. First, we estimated the long-wavelength height error by comparing SRTM against ICESat “centroid” elevation, and removed it by interpolating the height bias within 1/6 degree tiles. Second, we calculated the tree height bias by subtracting ICESat “lowest” elevation from SRTM. We estimated tree height bias in pixels without ICESat measurement using the Landsat tree density map and the global forest height map. Third, short-wavelength noise were reduced by a smoothing filter. By applying this 3-step method, the RMSE was also reduced from 8.06m to 5.65m. The modified DEM showed reasonable agreements to existing river network map and inundation map, suggesting the robustness of the removal method.

Journal

Keywords

Details 詳細情報について

  • CRID
    1390001205711951232
  • NII Article ID
    130005175941
  • DOI
    10.11520/jshwr.29.0_100
  • Text Lang
    ja
  • Data Source
    • JaLC
    • CiNii Articles
  • Abstract License Flag
    Disallowed

Report a problem

Back to top