Creation of a coastal evolution prognostic model using shoreline historical data and techniques of digital image processing in a GIS environment for generating future scenarios

• Autor(a) principal: Sousa, Willamys R. N. de
• Data de Publicação: 2018
• Outros Autores: Souto, Michael V. S., Matos, Stefanny S., Duarte, Cynthia R., Salgueiro, Ana Rita Gonçalves Neves Lopes, Silva Neto, Cláudio A. da
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da Universidade Federal do Ceará (UFC)
• Texto Completo: http://www.repositorio.ufc.br/handle/riufc/64547
• topic: Coastal erosion; Remotesensing; Python programming language; OpenCV library; Geographic Information Systems; Coastal evolution prognostic model

The process of coastal erosion is a global problem that impacts approximately 70% of coastal regions of the Earth. It causes loss of property, infrastructure, and biodiversity, besides generating major economic impacts. Therefore, the analysis and monitoring of coastal erosion is an issue that needs to be addressed. In this sense, remotesensing data have been widely used in studies that evaluate the spatial and temporal changes of land use. In addition, the use of time series of satellite imagery applied in the investigation of changes in the Earth’s coverage and its spatio-temporal pattern has been proven as an extremely efficient approach. Thus, remote sensing and geoprocessing are effective techniques to obtain continuous and dynamic information from coastal regions at different levels and scales. In this context, the main objective of this work was to create a prognostic model for the generation of future scenarios, based on the analysis of the spatialtemporal changes of the shorelines from past decades to the present, having as the pilot area the coast of the municipality of Icapuí, in the State of Ceará, Northeastern Brazil. For that, Statistical Regression technique was used. In addition, the techniques of Digital Image Processing and the extraction of the modified normalized difference water index were used. As a result, the prognosis of coastal erosion was generated for the year 2021, based on the time series of the years 1985, 1991, 1997, 2003, 2009, and 2015. After the extrapolation process, the results were validated through the mean absolute error. Furthermore, through the Python programming language and the OpenCV library, a computational solution was implemented to be executed in a Geographic Information Systems environment that automated the process of generating future prognostic and the extraction of the shoreline in a shapefile format.