Analise multitemporal do espectro da água na planície de inundação da barragem da usina hidrelétrica de Porto Primavera, São Paulo

Abstract

The Paraná River is the main river of the Prata Basin and is born from the confluence of the Grande and Paranaíba rivers with springs in the State of Minas Gerais and is the tenth largest river in the world in discharge and also the fourth in a drainage area. Along the high course of the Paraná River there are numerous physical impediments, characterized as dams, except the segment between the mouth of the Paranapanema River and the city of Guairá, where the river presents natural conditions. The hydrological and geomorphological character studies are important só that there is an analysis and understanding of the great number of problems, among them there is the supply, navigation, control of floods, among others. The present study is justified before the research tools such as geoprocessing and remote sensing, of which one has great relevance, since they are important helpers in the understanding of processes that happen in the reservoir. The present work aimed to analyze the multitemporal spectral behavior of water in the floodplain of the Paraná River downstream of the construction of the dam at the Porto Primavera Power Plant, São Paulo, qualitatively identifying the flood area of the Porto Primavera Plant, with CBERS 4 and Resourcesat 1, in time series from 2004 to 2017, validating the Near Infrared Spectrum (IVP) spectrum in images of the CCD – CBERS 2 and 2B sensors, MUX – CBERS 4 And the LISS3 sensor from Resourcesat in water recognition (reflectance and / or energy absorption) and monitoring the seasonality of the spring season’s climate, according to the spectral behavior of the water. The satellite images of the flood area were collected from the INPE database, the period being analyzed during the spring season. The images were processed in the software Spring 5.5.0, where they were firstly contrasted and cut to the limit of the desired area, later the pixel reading of the areas where the water was present, and finally being remapped with the appropriate intervals of pixel. The results obtained were representative figures of the area of the flood plain, highlighting its wetter areas. It was concluded that the research presented a broader view of seasonality in the climate of the same season in the floodplain of the dam, and it can be shown that according to the year, the climate changes are visibly clear. The years of greatest rainfall were 2007, 2009 and 2010 and the driest year was 2006. Completing also that the near infrared band met the expectations of the present research. For future study, with the same parameters and objectives, it is possible to use images from other satellites that have a better quality for the area of flood plain, being better visualized the characteristics of both the plain and the vegetation to its surroundings.