Revista Brasileira de Cartografia (Dec 2006)

Assessment of altimetry data in Amazonian forest based in INSAR, LIDAR; and GPS technologies.

  • Corina da Costa Freitas,
  • João Roberto Santos,
  • José Cláudio Mura,
  • Luciano Vieira Dutra,
  • Marcos Antônio Timbó Elmiro

Journal volume & issue
Vol. 58, no. 3
pp. 233 – 246

Abstract

Read online

There is a great amount of altimetry data collected by several experiments taken in the Tapajós National Forest located in the Brazilian Amazon State of Pará. Some of these data were produced by current state of the art technology whose effectiveness is still being proven by scientific investigations. In 1999 LIDAR profiles associated to videography data were taken in the region. In 2000 it was collected P and X band interferometric multifrequency data over areas of different vegetation types. The backscatter response collected by P band radar antenna potentially produces a real digital terrain model (DTM) due to its penetration capability across forest canopy toward the soil. When collected in X band radiation is reflected in the top forest canopy which produces a digital surface model (DSM). The availability of digital terrain model and digital surface model covering the same forested area is of great interest for many purposes, especially cartographic applications and carbon stock estimation. Also in 2000, the SRTM mission provided global interferometric cover in C and X bands that permitted the use of these data in the present work. In 2001, 2002, 2003 e 2005 several field control positions were surveyed using GPS dual frequency receivers and by topography surveying methods using total stations. The purpose of this work is to make an evaluation of the errors affecting the original multiple sensor data collected in the study area and apply an effective correction in order to eliminate those errors to permit using the models in other applications. An effective correction type was developed that reduced the original errors. The correction methodology consisted in altimetry offset determination and its application to the original data. Evaluations confirmed that the correction methodology produced good results and the corrected models presented global and local errors lesser than those in the original models.

Keywords