Evaporação em reservatórios do Nordeste brasileiro: avaliação da distribuição espacial e influência da mata ripária por sensoriamento remoto

• Ano de defesa: 2020
• Autor(a) principal: Rodrigues, Italo Sampaio
• Orientador(a): Costa, Carlos Alexandre Gomes
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Evaporação em reservatórios; Hidrologia; Sensoriamento remoto; Imagens Landsat; Modelos de evaporação; Evaporation in reservoirs; Hydrology; Remote sensing; Landsat images; Evaporation models
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/53463

Evaporation is one of the main hydrological processes in reservoirs in the Northeast of Brazil, and its quantification is still very uncertain. Therefore, the aim of this research was to assess the spatial distribution of evaporation in eight reservoirs, located in sub-humid and semi-arid climate, in the Brazilian Northeast using remote sensing. Scenes from Landsat 5 and Landsat 8 satellites were used fusing with four evaporation models: i) Linacre (1977); ii) Linacre (1993); iii) Blaney and Criddle (1950); and iv) AquaSEBS. For reference evaporation the Class A Pan and Piché Evaporimeter methods closest to the reservoirs were considered. The riparian vegetation was associated with the vegetation indices - NDVI and its influence on evaporation was spatially evaluated. As results it was obtained that, among the evaluated models, AquaSEBS was the one that presented the best average performance (NSE 0,27 – 0,58; PBIAS -17,00% - 5,20%; RMSE 0,81 – 1,17 mm/day) for most of the reservoirs. It was observed that evaporation is not homogeneous over the reservoir surface, with considerable spatial variability, with larger evaporation at the margins and on the region of the dam wall. This evaporation difference may be a consequence of direct water contact with the dam wall, receiving an extra amount of thermal energy from the soil. However, areas near the riparian forest presented lower evaporation rates, reductions between 18,4% a 30,7%, than the average. This interdependence was evidenced by the high correlation (R²: 0,87 – 0,96) between NDVI and evaporation, revealing the importance of riparian vegetation in reducing surface evaporation, a consequence of shading and the physical barrier caused by vegetation on the reservoir margin, reducing radiation and local aerodynamic influence, respectively. Depending on the spatial variability of evaporation, it was found that the volumes transferred to the atmosphere may have differences of up to 30%, which may vary according to the evaporation pixel chosen on the surface of the reservoir. On average, the evaporated volume in the eight reservoirs would be enough to supply more than two million people per day.