SURVEYING THE TOPOGRAPHIC HEIGHT FROM SRTM DATA FOR CANOPY MAPPING IN THE BRAZILIAN PANTANAL
Abstract
An algorithm was developed in Geographical Information Systems (GIS) for the extraction of topographic height from the Digital Elevation Models (DEM) of the Shuttle Radar Topography Mission (SRTM), C and X bands, applied to mapping vegetation canopy in the Pantanal Floodlands. According to previous studies, these bands are sensitive to surface vegetation and thus elevation values increase in relation to terrain proportional to the height of the canopy, known as canopy effect. The proposed algorithm identifies minimum elevations within a search radius, which are likely to represent bare earth values, to generate a reference surface. The topographic height results from the subtraction between the elevations of this surface and of the original DEM. It is expected that, in this region, the height values be related to the vegetation height. Whenever possible, the algorithm results were observed together with optical images and vegetation maps of RADAMBRASIL project, for the establishment of height slicing levels as related to vegetal groups. The visual examination and statistical analysis have provided three levels of slicing height, which would be related to herbaceous, shrub and tree (forest) vegetation communities. Although slicing levels could be related to general classes of vegetation canopy in this region, field data and, or, fine resolution optical data are required for more detailed mappings. Regardless of the classification approach, height estimates from SRTM DEM represent a subsidiary data for the remote characterization of the Pantanal Floodlands vegetation, which complements traditionally used optical dataDownloads
Published
Issue
Section
License
The authors maintain the copyright and grant GEOGRAFIA the right of first publication, with the articles simultaneously licensed under the Creative Commons BY 4.0 License, which allows sharing and adapting the articles for any purpose, as long as appropriate credits and provisions of image rights, privacy or moral rights. Other legal attributions can be accessed at: https://creativecommons.org/licenses/by/4.0/legalcode.en.
Geography, Rio Claro, SP, Brazil - eISSN 1983-8700 is licensed under the Creative Commons BY 4.0 License.