Estimativa de armazenamento de água em reservatórios através de sensoriamento remoto

• Ano de defesa: 2019
• Autor(a) principal: Flores, Yesica Ramirez
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• dARK ID: ark:/26339/0013000016r49
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Engenharia Agrícola; UFSM; Programa de Pós-Graduação em Engenharia Agrícola; Centro de Ciências Rurais
• Palavras-chave em Português: Disponibilidade hídrica; Imagem multiespectral; MDE; TIN; Water availability; Multispectral image; CNPQ::CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA
• Link de acesso: http://repositorio.ufsm.br/handle/1/17248

Freshwater reservoirs provide essential services to the population by providing drinking water for domestic, industrial, and agricultural use. In this way, finding appropriate means to manage this resource is vital for today's society. However, the collection of this information is an obstacle to high collection costs and difficult access to troubled areas. However, the use of remote sensing tools is potentially applicable for detection and monitoring of water bodies, overcoming the aforementioned problems. The objective of this work was to evaluate the use of remote sensing tools in the identification and determination of storage capacity of fresh water reservoirs for the municipalities of Alegrete, Itaqui, Quaraí and Uruguaiana in the State of Rio Grande do Sul, Brazil. Aiming at reaching the objectives proposed in the work, three situations were chosen for the choice of multispectral images, being these droughts, floods and periods of intense water use. Data from the OLI and MSI sensors (Landsat 8 and Sentinel-2B respectively) were analyzed and classified using ERDAS 2014® Software where by means of a 3x3 Laplacian filter the edge enhancement was facilitated, facilitating the delimitation of the reservoirs. The ArcScan tool in ArcGIS 10.4.1 for Desktop Advanced software was used for vectoring the targets. In contrast, elevation models SRTM and ALOS-PALSAR were also used for the identification / detection of the targets. Afterwards, the identified reservoirs were characterized, and it was possible to calculate the variables Area, Volume and Height of the reservoirs. Volumetric capacity and water availability were obtained for the study area. For the method of automatic identification of reservoirs using multispectral images, the data obtained with the Sentinel 2B-MSI image was the one that presented the best result for the study area, allowing the identification of 1.517 reservoirs, facilitating the identification of smaller reservoirs compared to Landsat 8 images - OLI. The ALOS PALSAR model identified 1.510 reservoirs for the study area, together with the identification of smaller reservoirs, compared to the SRTM model. The mean volume available in the study area was 2.5 billion cubic meters. The results obtained demonstrate the potential of the use of these remote sensing tools in the identification and characterization of water resources for the various purposes. Providing a useful tool to quantify available water in reservoirs, allowing managers, whether linked to agriculture or urban demand, to manage this resource in the best possible way.