ESTIMAÇÃO DA LINHA DE COSTA, COM USO DE IMAGENS ORBITAIS E DADOS MAREGRÁFICOS, PARA PRODUÇÃO/ATUALIZAÇÃO DE PRODUTOS CARTOGRÁFICOS NÁUTICOS

Abstract

Brazil has coastlines that are 10,959 km long. DHN is the national hydrographic service responsible for analyzing and representing the coastline (LC) in Brazilian nautical charts. This LC is a dynamic phenomenon whose study is of interest to several professionals in the field of Earth sciences, and the representation of this LC in Brazilian nautical charts, referred to the vertical indicative Mean High Water Springs (MHWS). The model for estimating the High Water Springs Line (LPS), related to MHWS, is based on the geometric relationship between the local tide height and its horizontal projection seen from space and recorded in satellite images, here called Tidal Projection (PM) model. Thus, we implemented the technique for the extraction of instantaneous coastlines (LCI) that is executed from QGIS software and developed the ANALICOS (coast line analysis), a script designed for data preparation, estimation and statistical analysis, implemented by the author in the Python programming language . From the input data, it was possible to verify the geometric coherence between the horizontal distances measured in the Sentinel-2 images with the local tidal height. Two references were used: the mark left by the water in a Sentinel image taken in the period of the sizigy tide and the coastline published by NOAA (National Oceanic and Atmospheric Administration) referred to the average high water line. As a result, the LPS estimated by the PM model was closer to the reference line extracted from the Sentinel image. Finally, the product was validated for use in nautical charts with scale from 1:25,000 (Class D), through an analysis by the Brazilian Cartographic Accuracy Standard (PEC-PCD) and analyzed under the criteria of the International Hydrographic Organization (IHO).