METODOLOGIA PARA PRÉ-PROCESSAMENTO DE MODELOS DIGITAIS DE ELEVAÇÃO

Abstract

The dissemination of altimetric data, through Digital Elevation Models (DEM), allowed the advancement of several scientific areas, more specifically Geomorphometry. However, the DEM may present noises, compromising the final results. This paper aims to propose a methodology for DEM pre-processing that optimises the subsequent process of generating morphometric variables. The proposed methodology consisted of: a) altimetric quality control; b) filling of spurious depressions; and c) applying filters. Among the evaluated DEM, the NASADEM obtained the best altimetric precision. The Fill Sinks algorithm allowed better removal of spurious depressions and the median low-pass filter, with moving window of 7x7 pixels allowed better identification of the relief forms of the study area. The methods proved to be suitable for study areas that present geomorphological heterogeneity.