Avaliação dos interpoladores krigagem e Topo to Raster na geração de Modelos Digitais de Elevação a partir de dados batimétricos

• Autor(a) principal: Carmo, Edilson José do
• Data de Publicação: 2014
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/3844
• topic: Ecobatímetro; Mapas batimétricos; Topo to Raster; Krigagem; Echo sounder; Bathymetric maps; Top to Raster; kriging; CNPQ::CIENCIAS EXATAS E DA TERRA::GEOCIENCIAS::GEODESIA

With developments in recent years, methods for bathymetric surveys using acoustic sensors (echo sounders) and signals transmitted by satellite navigation receivers, it becomes possible to describe the submerged same level of detail which is embossed with the surface relief describes terrestrial. A graphical representation of the underwater relief occurs from Digital Elevation Models (DEMs) generated by interpolating seeking, from measurements, predict the depth at unsampled locations. The geometrical information taken from bathymetric surveys, the volume of liquid water or mud present, for example in a reservoir, is the most relevant. This study sought to determine the products generated by single beam bathymetric survey, as well as the methods of kriging interpolation and Topo to Raster. The study areas were a box of decanting of the water treatment plant and a dam lake of Furnas, where surveys were conducted, with total station and bathymetric surveys, with a single beam echo sounder which operates at frequencies 33 kHz and 210 kHz . In the settling box, first bathymetric survey was conducted (Lbat 1). Then be emptied and cleaned the box and before filling it again, surveying, (Ltop) was performed with total station. Immediately after filling, without water and mud decanted, another bathymetric survey (Lbat 2) was performed. To survey the dam lake of Furnas, held automated bathymetric survey, ie, using the GNSS technology for positioning. First, the data from the survey, we evaluated the methods of kriging interpolation and Topo to Raster in the DEM generation of the settling box. The conclusion was that the interpolator Top to Raster conditioning showed remarkable deformations at the edges and in the center of the study area and should be discarded. The next stage of this work was to evaluate the accuracy of DEMs generated by applying kriging on data from single beam bathymetric survey, Lbat 2, using the frequencies of 33 kHz and 210 kHz. Possession of DEMs were calculated discrepancies between information from them and points of surveying. The results showed an accuracy of about 5 cm in mean depths of 3.21 m, and that as the surveys were conducted after cleaning the box, with clean water, there was no significant difference in the accuracy of DEMs generated with depths raised to the frequencies 33 kHz and 210 kHz. Then the DEMs were generated by kriging for the first bathymetric survey, Lbat 1, when solid waste resulting from the settling process still existed. Volumes were calculated and compared to evaluate the frequency of 33 kHz echo sounder to determine the volume of mud in the box. After analyzing the results, it was found that using only the first bathymetric survey, with the frequencies 33 kHz and 210 kHz, there was detected 186 m3 of sludge, in a total volume of 799 m3 of water and mud which, with the clean water the volume is determined by accurate bathymetry, showing discrepancy of 0.63% for the frequency of 210 kHz and 0.12% for the frequency of 33 kHz. To survey the lake of Furnas DEMs were generated by kriging interpolation method and the Top to Raster, varying the spacing between the regular survey lines using horizontal scan lines to the regular survey lines. Through statistical analysis of the discrepancies between estimated by interpolating depths and depths observed in the scan lines of the bathymetric survey, the kriging showed better results for spacings of 40 and 80 meters. Considering the isobatimétricas between different DEMs created, it is observed that for larger spacings the interpolator to Top Raster presented smoother features when compared to the DEMs generated by kriging.